SJ9HU ouqej
jpj |BnuB|Ai
90UBU9iU]B|/\|
pue uoiiBJQdo
030/98-l/SZ9/Vd3
iuaujdo|SA8a PUB q^essey
-------
EPA/625/1-86/020
June 1986
Manual
Operation and Maintenance
Manual for
Fabric Filters
by
PEI Associates, Inc.
11499 Chester Road
Cincinnati, Ohio 45246
Contract No. 68-02-3919
Project No. 3587
EPA Project Officer
Louis S. Hovis
Gas Cleaning Division
Paniculate Technology Branch
Air and Energy Engineering Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
11
-------
CONTENTS
Figures v
Tables viii
Acknowledgment t. ix
1. Introduction 1-1
1.1 Scope and content 1-2
1.2 Intended use of manual 1-3
2. Overview of Fabric Filter Theory, Design, and O&M Considerations 2-1
2.1 Basic theory and principles of filtration 2-1
2.2 Fabric filter systems 2-8
2.3 System components 2-17
2.4 Fabric filter O&M considerations 2-30
References for Section 2 2-39
3. Fabric Filter Performance Monitoring 3-1
3.1 Key operating parameters and their measurement 3-1
3.2 Instrumentation systems and components 3-10
3.3 Performance tests and parameter monitoring 3-12
3.4 Recordkeeping practices and procedures 3-18
4. Performance Evaluation, Problem Diagnosis, and Problem Solutions 4-1
4.1 Performance evaluation 4-1
4.2 Data collection and compilation 4-2
4.3 Problem diagnosis 4-6
4.4 Corrective actions 4-26
5. O&M Practices 5-1
5.1 Operating procedures 5-1
5.2 Preventive maintenance practices 5-9
References for Section 5 5-17
m
-------
CONTENTS (continued)
6. Inspection Methods and Procedures 6-1
6.1 Prestartup inspections 6-1
6.2 Startup inspection procedures 6-2
6.3 Routine preventive maintenance inspections 6-4
6.4 Diagnostic inspections 6-9
6.5 Recordkeeping 6-17
6.6 Summary 6-17
References for Section 6 6-21
7. Safety 7-1
7.1 Hopper entry 7-1
7.2 Confined area entry 7-3
7.3 Worker protection 7-9
8. Model O&M Plan 8-1
8.1 Management and staff 8-2
8.2 Maintenance manuals 8-5
8.3 Operating manuals 8-7
8.4 Spare parts 8-9
8.5 Work order systems 8-10
8.6 Computerized tracking 8-16
8.7 Procedures for handling malfunction 8-19
References for Section 8 8-20
Appendices
A. Examples of Fabric Filter O&M Forms A-l
B. Operation and maintenance of utility fabric filters B-l
Glossary G-l
-------
FIGURES
Number Page
2-1 Initial Mechanisms of Fabric Filtration 2-2
2-2 Dust Accumulation on Woven Glass Fabrics 2-4
2-3 Small Shaker-type Fabric Filter 2-9
2-4 Example of a Small Cylindrical Reverse-air Fabric Filter
Design 2-11
2-5 An Example of a Large Reverse-air Fabric Filter 2-12
2-6 Example of a Small Pulse-jet Fabric Filter 2-14
2-7 Typical Fabric Weaves 2-18
2-8 Cross Section of a Thimble Protecting Bottom of Bag 2-27
2-9 Methods of Bag Attachment in Shaker and Reverse-air
Fabric Filters 2-29
2-10 Proper Method of Installing Bag in Tube Sheet With Snap
Rings 2-31
2-11 Typical Spring Bag Tensioning Arrangement 2-32
2-12 Tubesheet Leak Due to Poor Welding at Wall of Pulse-jet
Fabric Filter 2-34
2-13 Broken Weld on Roof of Fabric Filter 2-35
3-1 Typical Cascade Impactor System 3-5
3-2 Sampling Train With Cascade Impactor 3-7
3-3 Example of a Properly Tensioned Bag That is Collapsed
During Reverse-air Cleaning 3-9
4-1 Typical Bag Replacement Record 4-5
-------
FIGURES (continued)
Page
4-2 Example of Exceptionally Poor Access to the Top of a
Pulse-jet Fabric Filter 4-8
4-3 Tied-off Fiberglass Bags During Bag Replacement 4-9
4-4 Example of a Diffuser for Reflecting Large Particles
From the Gas Stream 4-14
4-5 Fugitive Dust Emissions Resulting From the Opening of
Hopper Access Doors to Clean Out Clogged Hoppers 4-16
4-6 Example of Bag Wear Caused by Corrosion of Metal Anti-
collapse Rings 4-21
4-7 Electronic Timer Circuit Board for a Pulse-jet Filter 4-24
4-8 Pulse-jet Solenoids for Individual Rows of Bags 4-25
4-9 Example of Correct and Incorrect Methods of Installing
Bags in a Reverse Air Fabric Filter " 4-28
4-10 Diaphragm Assembly for Pulse-jet System With Solenoid
Connection Removed 4-37
4-11 Misaligned Blow Pipe in Pulse-jet Fabric Filter Caused
By a Broken Bolt at the End of the Pipe 4-40
5-1 Precoating Material for Protection of Bags From Blinding 5-3
5-2 Limestone Precoat Hardened Because of Condensation
Problems 5-5
5-3 Examples of Strip Chart Output on a Fabric Filter 5-8
5-4 Use of an Ultraviolet Light to Check for Leaks of Fluores-
cent Dye that Has Been Injected into the Fabric Filter 5-15
6-1 Cleaning Valve Problems 6-6
6-2 Pinhole Leaks in Bags Can Be Determined By Watching for
Emissions Immediately After the Cleaning Pulse Is Fired 6-7
6-3 Knocking Dust Off Bags to See How Easily Removable It Is 6-10
6-4 Bag-to-bag Contact in a Pulse-jet Fabric Filter Resulting
From Poor Alignment of Cages During Installation 6-12
VI
-------
FIGURES (continued)
Page
6-5 Common Bag Problems With Pulse-jet Fabric Filters 6-13
6-6 Pinhole Leak Forms a Dust Jet on the Floor Near a Shaker-
type Fabric Filter 6-15
6-7 Checking For Excessive Build-up and Poorly Tensioned Bags
in a Reverse-air Fabric Filter 6-16
6-8 Routine and Diagnostic Inspection Data for Pulse-jet
Fabric Filters 6-19
6-9 Routine and Diagnostic Inspection Data for Reverse Air
and Shaker Fabric Filters 6-20
7-1 Nomograph Developed by McKarns and Brief Incorporating
the Revised Fort Knox Coefficients 7-14
8-1 Organizational Chart for Centrally Coordinated Fabric
Filter O&M Program 8-4
8-2 Outline For Fabric Filter Maintenance Manual 8-6
8-3 Fabric Filter Operating Manual Outline 8-8
8-4 Example of Five-level Priority System 8-13
8-5a Example of Work Order Form 8-17
8-5b Example of Work Order Form 8-18
vii
-------
TABLES
Number Page
2-1 Recommended Temperature Limits for Various Fabrics 2-21
2-2 Chemical Resistance of Common Commercial Fabrics 2-23
5-1 Typical Maintenance Inspection Schedule for a Fabric
Filter System 5-11
7-1 Effects of Various Levels of Oxygen on Persons 7-5
7-2 Allowable Concentrations for Entry Into Confined Spaces 7-6
7-3 Applications Presenting Potential Eye Hazards 7-9
7-4 Maximum Permissible Sound Level for Intermittent Noise 7-10
7-5 Threshold Limit Values for Nonimpulsive Noise 7-11
7-6 Index of Heat Stress 7-13
7-7 Heat Production for Various Levels of Exertion 7-15
7-8 Body Heat Production as a Function of Activity 7-16
-------
ACKNOWLEDGMENT
This manual was prepared for the U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory under contract No. 68-02-3919.
Mr. Michael F. Szabo was the project manager and provided technical coordina-
tion/editing in the preparation of this manual. The primary authors of this
manual were Fred D. Hall, Gary L. Saunders, Ronald L. Hawks, and Michael F.
Szabo. Other contributing authors were David R. Dunbar and Michael T. Melia.
Mr. Jack A. Wunderle provided senior review and assisted in developing the
detailed report outline. Ms. Marty H. Phillips provided editorial services
and the page layout design. Mr. Jerry Day coordinated typing and graphics
for the report and provided final review.
A review panel consisting of 34 members provided input throughout .this
project. They are -listed below in alphabetical order:
Affiliation
Name
Charles A. Altin
Ralph Altman
William S. Becker
Eli Bell
William S. Bellanger
Robert Brown
Steven Burgert
John M. Clouse
Jim Cummings
Duane Durst
Heinz Engelbrecht
David Ensor/David Coy
Kirk Foster
F.W. Giaccone
Wally Hadder
James Hambright
Norman Kulujian
Ebasco Services, Inc.
Electric Power Research Institute
State and Territorial Air Pollution Program
Administrator
Texas Air Control Board
U.S. EPA Region III
Environmental Elements Co.
East Penn Manufacturing Co.
Colorado Department of Health
U.S. EPA Office of Policy Analysis
U.S. EPA Region VII
Wheelabrator-Frye
Research Triangle Institute
U.S. EPA Stationary Source Enforcement
Division
U.S. EPA Region II
Virginia Electric Power Company
Pennsylvania Bureau of Air Quality Control
U.S. EPA Center for Environmental Research
Information
IX
-------
Name Affiliation
John Lytle Tennessee Valley Authority
Richard McRanie Southern Company Services
Grady Nichols Southern Research Institute
Sid Orem Industrial Gas Cleaning Institute
John Paul Montgomery County, Ohio Regional Air
Pollution Control Agency
Charles Pratt U.S. EPA - Training
Richard Renninger National Crushed Stone Association
John Richards Richards Engineering
A.C. Schneeberger Portland Cement Association
Eugene J. Sciassia Erie County, New York Department of Envi-
ronment and Planning
Don Shephard Virginia Air Pollution Control Board
Lon Torrez U.S. EPA Region V
William Voshell U.S. EPA Region IV
Glenn Wood Weyerhauser Corp.
Howard Wright U.S. EPA Stationary Source Enforcement
Division
Earl Young American Iron and Steel Institute
The comments of the review panel on the topics to be covered, the de-
tailed outline of the manual, and the draft manual were very helpful and have
contributed to the success of this project.
Finally, the cooperation and assistance of the project officer, Mr.
Louis S. Hovis, in completing this manual is greatly appreciated.
-------
SECTION '
INTRODUCTION
The success of an air pollution abatement program ultimately depends
upon effective operation and maintenance (O&M) of the installed air pollution
control equipment. Regardless of how well an air pollution control system is
designed, poor O&M will lead to the deterioration of its various components
and a resulting decrease in its particulate removal efficiency. This is
particularly true with fabric filters; if not corrected, a small problem like
a pinhole leak in a bag can result in the failure of the surrounding bags
within a short time.
Effective O&M also affects equipment reliability, on-line availability,
continuing regulatory compliance, and regulatory agency/source relations.
Lack of timely and proper O&M leads to a gradual deterioration in the equip-
ment, which in turn increases the probability of equipment failure and de-
creases both the reliability and on-line availability of the equipment.
These latter two items can decrease plant productivity if process operations
are forced to be curtailed or shut down to minimize emissions during air
pollution control equipment outages. Frequent violations of emission limits
can result in more inspections, potential fines for noncompliance, and in
some cases, mandatory shutdown until emission problems are solved.
This manual focuses on the operation and maintenance of typical fabric
filters. The overview presented in Section 2 summarizes the available infor-
mation on fabric filter theory and design in sufficient detail to provide a
basic background for the O&M portions of the manual. Numerous documents are
available if the reader desires a more rigorous treatment of fabric filter
theory and design.
The manual is designed to be an educational tool for plant and EPA
personnel, not an enforcement tool. No attempt is made to tell plant per-
sonnel how to operate a plant; rather, the manual provides cause-effect
SECTION 1-INTRODUCTION
-------
relationships to assist in the prevention or location of problems. Actual
plant experience and examples are used to demonstrate or illustrate certain
points. The manual will not only serve as a handy reference regarding fabric
filter O&M, but also provide the necessary information to assist plant per-
sonnel in the preparation of their own site-specific O&M manual.
1.1 SCOPE AND CONTENT
Section 2 of the manual presents an overview of fabric filter theory and
design in sufficient detail to provide a background for the other sections in
the manual. Although this section does not provide "textbook" coverage of
fabric filter theory and principles (such information is readily available in
the technical literature), it does draw attention to some new information or
recent developments that may be useful in improving fabric filter O&M. This
section also presents a general description of fabric filter types and exam-
ines their applicability to varying gas stream conditions and particulate
characteristics.
Section 3 presents the purpose, goals, and role of performance monitor-
ing as a major element in an O&M program. Discussions w.ill focus on key
performance indicators and their measurement and on instrumentation, data
acquisition, and recordkeeping methods useful in optimizing fabric filter
system performance.
Section 4 discusses the use of performance monitoring and other informa-
tion to evaluate equipment performance, diagnose real or impending problems,
and troubleshoot problem and malfunction causes. Tracking procedures and
trends analysis methods that can be used to assess current or impending dete-
rioration in performance are also discussed. Problem diagnosis and potential
corrective measures are described.
Section 5 presents guidelines for general O&M practices and procedures
for use in improving and sustaining fabric filter performance and reliabili-
ty. General guidance, rather than specific instructions, is given because of
the unique nature of these control device systems and of the process streams
they serve. The intent of this section is to prescribe the basic elements of
good operating practice and preventive maintenance programs that can be used
as the basis and framework for tailored, installation-specific programs.
SECTION 1-INTRODUCTION
-------
Section 6 presents methods and procedures for detailed inspections of
fabric filter systems and their components. It provides step-by-step proce-
dures and techniques for conducting external "and internal inspections at both
large and small fabric filter installations. Inspection during the pre-
operational construction phase and the performance demonstration (baselining)
period are also addressed. Safety considerations are emphasized and any
special precautionary measures at major source installations are stated. The
portable instrumentation and safety equipment needed during inspection are
listed, and example inspection checklists are provided.
Section 7 summarizes important items that an adequate O&M plan should
include. Different plans are suggested for small vs. large fabric filter
operations. The format of this section consists of an annotated outline of
the key items that should be included in a model O&M plan.
Appendix A shows examples of fabric filter O&M forms. Appendix B summa-
rizes the development and status of fabric filter application to utility
boilers. Although this is an application that is still being developed, many
of the initial problems have already been solved. A glossary of terminology
is also presented.
1.2 INTENDED USE OF MANUAL
The intent of this manual is to present O&M principles and procedures to
supplement plant-specific O&M techniques and strategies. An underlying
objective is to improve air quality by providing technical guidance that will
upgrade the reliability and performance of fabric filters.
The intended audience is the plant environmental engineer, plant O&M
personnel, and EPA field personnel. The contents are slanted toward the con-
cerns of the plant environmental engineer, however, who with the assistance
of his/her staff is responsible for long-term control strategies, O&M plans,
preparation of bid specifications, and performance trends analyses. The doc-
ument also presents information to enable plant O&M personnel to recognize
potential problem areas as well as existing problems, their underlying caus-
es, and their solutions. The information provided should help EPA field per-
sonnel to determine if the fabric filter is operating within the applicable
SECTION 1-INTRODUCTION 1_3
-------
regulations, to judge the effectiveness of the plant's O&M program, and to
assess the causes of poor fabric filter performance.
The plant environmental engineer will generally have overall responsi-
bility for the proper O&M of the fabric filter, for a wide range of trends
analyses related to performance of the fabric filter, and for background
information during the preparation of bid specifications. This manual does
not attempt to replace the step-by-step O&M manuals prepared by fabric filter
vendors or documents developed by the plant for a site-specific application.
Neither does it directly address the development of bid specifications. It
does, however, attempt to provide sufficient detail to enable the plant envi-
ronmental engineer to evaluate the plant's present O&M program and determine
if and where improvements are needed. The examples presented throughout the
text can be used to become aware of potential problems that can occur that
the engineer may not have experienced as yet. The descriptions, general
procedures, and analytical techniques presented can assist the plant environ-
mental engineer and the central engineering staff in developing bid specifi-
cations for purchase of a fabric filter that will facilitate O&M as well as
performance evaluations and trends analyses.
Plant O&M personnel should not use this manual for specific instructions
on startup and shutdown procedures. Such instructions should be provided by
the equipment manufacturer, and minor modifications should be made by plant
personnel to fit site-specific needs. This manual presents general operating
guidelines that can be used as a background document for determining the com-
pleteness of the plant's fabric filter operating manual, preventive and cor-
rective maintenance procedures, and troubleshooting and inspection procedures,
For EPA field personnel, the manual provides guidelines for a detailed
field inspection of fabric filter systems. Emphasis is on the inspection
methodology for evaluating both equipment and performance. Discussions do
not include topics covered in detail elsewhere (e.g., source testing and
opacity readings).
Other user groups will find the manual provides useful general informa-
tion on fabric filter O&M and problems encountered in a wide variety of fab-
ric filter applications. The manual clearly shows that actual fabric filter
O&M and performance often differ greatly from theorized O&M and performance.
SECTION 1-INTRODUCTION
1-4
-------
SECTION 2
OVERVIEW OF FABRIC FILTER THEORY,
DESIGN, AND O&M CONSIDERATIONS
The general descriptions of the basic theory and principles of filtra-
tion, fabric filter systems, and O&M considerations presented in this section
are intended only to provide a background for the remainder of the manual.
Detailed information is readily available in various textbooks and in the
proceedings of pertinent seminars and conferences.
2.1 BASIC THEORY AND PRINCIPLES OF FILTRATION
An understanding of the basic particle collection mechanisms and the
factors that affect fabric filtration are necessary for any evaluation of
parameter monitoring data for the establishment of optimum O&M procedures.
It is important in the analysis of the cause-effect relationships between
variations in operating parameters and operating conditions and fabric filter
performance.
2.1.1 Particle Capture Mechanisms
A single fiber can be used to describe the various capture mechanisms of
a fabric filter. As shown in Figure 2-1, the five basic mechanisms by which
particulate can be collected by a single fiber are 1) inertial impaction, 2)
Brownian diffusion, 3) direct interception, 4) electrostatic attraction, and
5) gravitational settling.
These collection mechanisms, plus sieving, also apply to a fabric with a
dust cake, such as would be encountered under typical operating conditions.
Inertia! impaction is the dominant collection mechanism within the dust cake.
The forward motion of the particles results in impaction on fibers or on
1 2
already deposited particles. ' Although impaction increases with higher gas
velocities, these higher velocities reduce the effectiveness of Brownian
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY, DESIGN, AND O&M CONSIDERATIONS 2-1
-------
DIRECT
INTERCEPTION
INERTIAL /
IMPACTION
ELECTROSTATIC
ATTRACTION
GRAVITATIONAL
SETTLING
Figure 2-1. Initial mechanisms of fabric filtration.
Reprinted with permission from Industrial Air Pollution Control Equipment for Particulates
by L. Theodore and A. J. Buonicore, Copyright CRC Press, Inc.
Boca Raton, Florida. 1S76.
ro
rvs
-------
diffusion. Increasing the fabric and dust cake porosity by use of a less
3
dense fabric or more frequent cleaning also reduces diffusional deposition.
Except at low gas velocities, gravity settling of particles as a method of
collection is usually assumed to be negligible.1' ' Electrostatic forces
may affect collection because of the difference in electrical charge between
the particles and the filter; however, the impact on commercial-scale equip-
ment is not fully understood. Sieving occurs when the particle is too large
to pass through the fabric matrix. It is not one of the major mechanisms for
collecting particulate. The combination of all these particle collection
mechanisms results in high-efficiency removal efficiency for all particle
sizes.
2.1.2 Dust Accumulation on Fabrics
The fabric filtration process or the accumulation of particulate on a
new fabric surface occurs in three phases: 1) early dust bridging of the
fabric substrate, 2) subsurface dust cake development, and 3) surface dust
cake development. The fabric used in a fabric filter is typically a woven or
felted material, which forms the base on which particulate emissions are col-
lected. Woven fabrics consist of parallel rows of yarns in a square array.
The open spaces between adjacent yarns are occupied by projecting fibers
called fibrils. Felted fabrics are constructed of close, randomly inter-
twined fibers that are compacted to provide fabric strength. Figure 2-2
depicts this particle accumulation on woven glass fabrics.
In the first phase, particles entering a new fabric initially contact
the individual fibers and fibrils and are collected by the filtration mecha-
nisms. These deposited particles, which are essentially lodged within the
fabric structure, promote the capture of additional particles. As these
particles build up during the second phase, particle aggregates form, bridg-
ing of the interweave and interstitial spaces occurs, and a more or less
continuous deposit is formed. In the third phase, particles continue to
collect on the previous deposit, and the surface dust cake is developed.
The cleaning cycle (via shaking, reverse air, or pulse jet) removes some
of the surface cake. After a few cleaning cycles, theoretically, a steady-
state dust cake should be formed, which will remain until the bag is damaged,
replaced, or washed. Actually, however, the dust cake can vary significantly
SECTION 2-OVERVIEW OP FABRIC FILTER THEORY. DESIGN. AND OlM CONSIDERATIONS
C- ""O
-------
DUST
YARN
YARN
BULKED FIBERS
UNUSED FABRIC
EARLY DUST BRIDGING OF FIBER SUBSTRATE
SUBSURFACE DUST CAKE DEVELOPMENT
SURFACE DUST CAKE DEVELOPMENT
Figure 2-2. Dust accumulation on woven glass fabrics.
10
-------
from cycle to cycle, particularly in severe applications such as utility
boilers or metallurgical processes. This remaining cake forms a base for the
collection of particles when the bag is put back on line after cleaning.
2.1.3 Gas Stream Factors That Affect Fabric Filter Design and Operation
A complete characterization of the effluent gas stream is important in
the design and operation of a fabric filter system. It should include the
gas flow rate; minimum and maximum gas temperatures; acid dew point; moisture
content; presence of large particulate matter; presence of sticky particulate
matter; particulate mass loading; chemical, adhesion, and abrasion properties
of the particulate; and presence of potentially explosive gases or particu-
late matter. These data can be used to design a collector with the required
degree of control or to optimize the operation of an existing fabric filter,
as illustrated by the following examples:
1) The size of a fabric filter system is determined by the gas volume
to be filtered, the air-to-cloth (A/C) ratio, and the pressure drop
at which the filter can be operated given the fabric type, dust
cake properties, and cleaning method. The area of fabric surface
is determined by multiplying the total gas flow by the selected A/C
ratio.
2) Penetration is related to the effective A/C ratio in the system,
particularly if the A/C ratio is outside the optimum range for the
specific application and type of fabric filter (see Section
2.2).9'10 Therefore, the lowest possible face velocity consistent
with economic constraints should be specified during the design
phase. This parameter should also be considered in the operation
of an existing fabric filter, if process flow rates increase sig-
nificantly or additional sources are added.
3) Variations in gas stream temperature over time affect the operation
and design of a fabric filter. The temperature of gases emitted
from industrial processes may vary more than several hundred de-
grees within short periods of time. They may fall below the gas
moisture and acid dew points or they may exceed the maximum that
the fabric will tolerate. The temperature extremes must be deter-
mined before the filter fabric is selected and during evaluation of
fabric filter performance.
4) The particle size distribution of the dust must be considered in
the design and operation of the collector. Particle size distribu-
tion affects both the porosity of the dust cake and abrasion of the
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O»M CONSIDERATIONS
-------
fabric. The presence of fine particles in the gas stream can cre-
ate a very compact dust cake and increase the static pressure drop
through the cake.11 These fine particles can also cause fabric
bleeding if pulled through the fabric. The presence of large
abrasive particles can reduce bag life and may necessitate the use
of a precleaner or gas distribution devices in the collection
system.
5) Moisture content and acid dew point are important gas composition
factors. Operating a fabric filter at close to the acid dew point
introduces substantial risk of corrosion, especially in localized
spots close to hatches, in dead air pockets, in hoppers, or in
areas adjacent to heat sinks, such as external supports.12'13
Allowing the operating temperature to drop below the water and/or
acid dew point, either during startup or at normal operation, will
usually cause blinding of the bags. Acids or alkali materials can
also weaken the fabric and shorten their useful life. Trace com-
ponents, such as fluorine, also can attack certain fabrics.
2.1.4 Fabric Filter Models
Fabric filter models have been developed to predict performance with
various operating and design parameters. The greatest drawbacks to fabric
filter models are their inability to account for real-world conditions, such
as acid dewpoint excursions, the rate of formation of pinhole leaks or bag
tears, and the effect of poor construction or maintenance. Modeling also
requires site-specific information that is not easily obtained, such as the K
factor (i.e., the specific resistance of the dust cake in inches water column
per pound of dust per square foot of filter area per feet per minute of
filtering velocity). In this section, a model developed by GCA Corporation
for the U.S. Environmental Protection Agency and applicable to coal-fired
boilers is discussed briefly to show the typical variables that are included
in a fabric filter model. Again, although none of the available fabric
filter models is very useful in the evaluation of the effect of O&M practices
on fabric filter efficiency, they can be used to some extent to evaluate the
effect of some design parameters on the theoretical efficiency of the unit.
14
The GCA model predicts the ability of fabric filters to clean coal-
fired boiler flue gases. This model addresses filters having either mechani-
cal or reverse-air cleaning mechanisms. It provides estimates of average and
point values for penetration and mass effluent concentration for a selected
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O&M CONSIDERATIONS
-------
set of operating parameters and dust/fabric specifications. The model ex-
presses penetration or outlet concentration as a function of the following
parameters:
Pn or CQ = f (*, C., W, F, CR)
where P = penetration
C = outlet concentration
$ = a parameter characterizing the dust/fabric combination of interest
C. = inlet concentration (constant)
CR = residual concentration
W = fabric dust loading
F = face velocity (determined by the air-to-cloth ratio)
To determine the parameter that influences fabric filter efficiency for
a specific process, one can investigate each of the parameters individually.
According to the model, face velocity is the design parameter with the great-
est impact on penetration. Under actual field conditions, however, other
factors may have a much more significant impact on fabric filter efficiency.
2.1.5 Fabric Filter Applications
The two fundamental applications of fabric filters are for nuisance dust
control and for process control. These applications are discussed briefly
below.
2.1.5.1 Nuisance Dust Control Applications--
Storage silos, woodworking shops, and materials transfer are examples of
the types of operations where fabric filters would be used to control dust in
the air. In many cases, low volumes of air are handled, and continuous
on-line cleaning, long-life fabrics, corrosion, and other factors are usually
not considerations in their design and operation.
2.1.5.2 Manufacturing Process Applications--
Fabric filter systems used to clean the exhaust from major processes may
be required to operate 24 hours a day, 365 days a year. The exhaust streams
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O&M CONSIDERATIONS
2-7
-------
may contain large high-temperature gas volumes, highly abrasive materials,
submicron particles, and corrosive agents. In some cases, the process fabric
filter may be critical to the plant's continued operation. In an asphalt
plant, for example, if the fabric filter is not operating properly, the plant
15
may not be able to achieve adequate production rates.
If toxic substances are present in the gas stream, operation of the
fabric filter is critical to the well-being of the workers and the general
public (e.g., in a plant where lead oxide is generated). For these reasons,
more sophisticated monitoring equipment (such as temperature and pressure
gages at an inside monitoring station rather than outside) is used to monitor
the performance of these units, and they generally represent a higher invest-
ment than fugitive dust collectors. Fabric filters controlling processes are
stressed in this manual because of the wide variation in operating conditions
to which they are exposed and the importance of troubleshooting and regular
maintenance for efficient operation.
2.2 FABRIC FILTER SYSTEMS
Although the basic particulate collection mechanisms are the same for
all fabric filters and gas stream factors affecting their performance are
relatively similar, the equipment itself and fabrics used in fabric filter
systems vary widely by vendor and application. Some of these variations are
necessary to meet various performance capability demands and physical charac-
teristics; others are the products of individual contributions of numerous
equipment and fabric vendors.
Although fabric filters can be classified in a number of ways, the most
common way is by their method of fabric cleaning: shaker, reverse-air, and
pulse-jet.
2.2.1 Shaker-Type Fabric Filters
A conventional shaker-type fabric filter is shown in Figure 2-3. Par-
ticulate-laden gas enters below the tube sheet and passes from the inside bag
surface to the outside surface. At regular intervals a portion of the dust
cake is removed by manual shaking (small systems) or mechanical shaking
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. OESION. AND OftM CONSIDERATIONS
L.-
-------
OVERMOUNTED
EXHAUSTER
DOOR-J
DOOR-
INLET
BAG -JV SHAKER
HANGER VNJT " "
\ CLOTH
A BAGS
BAG NOZZLE '
AND RETAINER
I
T
INLET CHAMBER
AND
HOPPER
BAFFLE
DISCHARGE GATE
Figure 2-3. Small shaker-type fabric filter
(Reprinted with permission from Flakt, Inc.)
2-9
-------
(larger systems). Mechanical shaking of the filter fabric is normally accom-
plished by rapid horizontal motion induced by a mechanical shaker bar at-
tached at the top of the bag. The shaking creates a standing wave in the bag
and causes flexing of the fabric. The flexing causes the dust cake to crack,
and portions are released from the fabric surface. The cleaning intensity is
controlled by bag tension and by the amplitude, frequency, and duration of
the shaking. Woven fabrics are generally used in shaker-type collectors; and
because of the low cleaning intensity, the gas flow is stopped before clean-
ing begins to eliminate particle reentrainment and to allow the release of
the dust cake. The cleaning may be done by bag, row, section, or compart-
ment.
Gas flow through shaker-type fabric filters is usually limited to a low
superficial velocity or A/C ratio of less than 3 ft/min and a typical range
of from 1 to 2 ft/min. High A/C values can lead to excessive particle pene-
tration or blinding, which reduces fabric life and results in high pressure
drop.
Mechanical shaker-type units differ with regard to the shaker assembly
design, bag length and arrangement, and type of fabric. All sizes of control
systems can use the shaker design.
2.2.2 Reverse-Air Fabric Filters
In fabric filters with reverse-air cleaning, particles can be collected
on a dust cake on either the inside or outside of the bag. Fabrics may be
either woven or felt, but felts are normally restricted to external surface
collection. A small cylindrical unit with external surface filtering is
illustrated in Figure 2-4. In this design, the bags are arranged radially
and are suspended from an upper tube sheet. The inner and outer row of
reverse-air manifolds continuously rotate around the unit and use a dampering
system to induce reverse flow in each bag. Thus, it is not necessary to
isolate the entire baghouse for dust-cake removal. A somewhat larger and
perhaps more typical reverse air filter is shown in Figure 2-5.
Regardless of design differences, the principle is the same. Cleaning
is accomplished by reversal of the gas flow through the filter media. The
change in direction causes the surface contour of the filter surface to
change (relax) and promotes dust-cake cracking. The flow of gas through the
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND OftM CONSIDERATIONS
-------
REVERSE AIR
PRESSURE BLOWER
OUTER ROW
REVERSE AIR
MANIFOLD
FABRIC FILTER
TUBES
INNER ROW
DRIVE REVERSE AIR CLEAN AIR
MOTOR MANIFOLD OUTLET
MIDDLE ROW
REVERSE AIR
MANIFOLD
PRE-CLEANING
BAFFLE
DUST AIR INLET
HEAVY DUST DROPOUT
Figure 2-4. Example of a small cylindrical reverse-air fabric filter design
(Courtesy of Carter-Day Company).
2-11
-------
Figure 2-5. An example of a large reverse-air fabric filter
(Courtesy of MikroPul Corporation).
2-12
-------
fabric assists in removal of the cake. The reverse flow may be supplied by
cleaned exhaust gases or by ambient air introduced by a secondary fan.
In filters with inside bag collection, cleaning is done with compart-
ments isolated. The filter bags may require anticollapse rings to prevent
closure of the tube and dust bridging.
Reverse-air filters are usually limited to A/C ratios of less than
3 ft/min, and a range of about 1 to 2.5 ft/min. In general, the appropriate
A/C ratio for a reverse-air unit should be about one-third lower than for a
similar shaker-type unit application.*
2.2.3 Pulse-Jet Fabric Filters
In pulse-jet fabric filters, filtering takes place on exterior bag
surfaces. A small pulse-jet fabric filter is illustrated in Figure 2-6. The
bags, supported by inner retainers (usually called cages), are suspended from
a tube sheet, an upper cell plate. Compressed air for cleaning is supplied
through a manifold-solenoid assembly into blow pipes. Venturis are mounted
in the bag entry area to improve the pulse-jet effect and to protect the top
part of the bag. The diffuser is placed at the gas inlet to prevent large
particles from abrading lower portions of the bag.
During cleaning, a brief (generally less than 0.2-second) pulse of com-
pressed air injected into the top of the bag creates a traveling wave in the
fabric, which shatters the cake and throws it from the surface of the fabric.
The dominant cleaning mechanism in a pulse-jet unit is fabric flexing. Felt-
ed fabrics are normally used, and the cleaning intensity (energy) is high.
The cleaning usually proceeds by rows, and all bags in a row are cleaned
simultaneously. The compressed-air pulse, which is delivered at 80 to 120
psi, results in local stoppage of the gas flow. The cleaning intensity is a
function of compressed-air pressure. Pulse-jet units can operate at substan-
tially higher A/C ratios than the previously discussed fabric filters because
•D
of their higher cleaning intensity. Typical ratios range from 5 to 10 ft /
2
ft -min.
*
Correspondence between Wheelabrator-Frye, Inc., and PEI Associates, Inc.,
February 1, 1985.
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN, AND O&M CONSIDERATIONS
2-13
-------
TUBE SHEET
CLEAN AIM PLENUM
PLENUM ACCESS
•LOW PIPE
INDUCED FLOW
BAG CUP & VENTURI
BAG RETAINER
TO CLEAN AIM OUTLET
AND EXHAUSTER
DIRTY AIM INLET » DIFFUSEM
Figure 2-6. Example of a small pulse-jet fabric filter
(Courtesy of George A. Rolfes Company).
2-14
-------
The plenum pulse cleaning method is a variation of the pulse-jet clean-
ing mechanism; in this method, an entire section of bags is pulsed with
compressed air from the clean air plenum.
2.2.4 Other Designs and Modifications
2.2.4.1 Positive-Pressure Vs. Negative-Pressure Units--
Fabric filters can be constructed either as a positive-pressure unit
with the fan upstream of the fabric filter or as a negative-pressure unit
with the fan downstream of the unit.
The use of a positive-pressure fabric filter eliminates the need for
ductwork and a stack downstream of the unit, which reduces requirements for
space and other materials but makes monitoring more difficult. Because
positive-pressure units are generally not exposed to as high a static pres-
sure as negative-pressure units, housing can sometimes be constructed of a
15
bolted light-gauge material. Any leaks from the fabric filter will enter
the surrounding air, which eliminates the potential cooling effects of unde-
sirable dilution air, but increases the risk of fugitive emissions from the
unit. Positive-pressure units are frequently used to control arc furnaces,
canopy exhausts, ferroalloy furnaces, and BOF reladling fumes.
In negative-pressure units, the fan is located on the clean side of the
filter, where it is subject to less wear from dust abrasion. The fabric
filter housing must be gas-tight, as any leaks will draw air in from the
outside. This outside air will normally cool the gas stream; and it could
reduce the gas temperature below the dew point, which would cause condensa-
tion on the inside of the unit. In some processes, introduction of outside
air increases the risk of fire and/or explosion. On the other hand, leaks
will not result in fugitive emissions because ambient air is drawn into the
unit. Negative-pressure units are generally used on coal-fired boilers,
cement kilns, rock dryers, carbon black reactors/dryers, and nonferrous
smelting furnaces.
2.2.4.2 Fabric Filter Control of Sulfur Dioxide--
Some fabric filters used to control boilers are designed or modified to
collect sulfur dioxide (SCL) with a dry SCL-collection system. The two basic
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O&M CONSIDERATIONS
2-15
-------
methods of dry S(L removal are 1) dry injection and 2) spray drying. The
first method consists of injecting a sodium compound (nahcolite) as a powder
into the flue gas upstream of the fabric filter. The dry sorbent collects on
the bags, and the S09 is removed, both in the suspended state and when the
17
gases pass through the cake that has built up on the bags. In the spray
drying technique, an atomized alkaline solution or slurry (lime) is injected
into the flue gas. As the alkaline material evaporates to dryness, it reacts
1 p
with the SO and HC1 in the flue gas. The resultant reaction products are
A
collected, usually along with the fly ash in the fabric filter, and disposed
of as a dry waste.
The primary advantages of both dry removal systems are the lower capital
cost associated with the simultaneous removal of SC^ and particulate in the
system, greater availability of the system because of its simplicity, and
lower energy and water consumption than with wet scrubbers. Spray drying
has an additional advantage in that it can be used in conjunction with either
fabric filters or electrostatic precipitators.
The main disadvantages of spray drying are that high removal efficien-
cies are more difficult to obtain, a highly reactive (i.e., nonlimestone)
absorbent must be used, and the spent sorbent cannot be easily disposed of.
2.2.5 Recent Developments and Trends
Many of the continuing efforts to develop new techniques, technologies,
and applications for fabric filters are not within the scope of this report
because they are either in the early stages of development or the topic is of
limited interest. These topics include:
0 Electrostatically enhanced filtration
0 Wet and granular filters
0 Grounded bags in coal mill installations
0 Counterweight bag tensioner
0 Stainless steel and other exotic materials for bags.
Further information on these topics can be obtained by reviewing specialty
conference proceedings and periodical articles.
The relatively new products that are described briefly include the
Staclean diffuser and sonic enhancement of bag cleaning. The Staclean dif-
fuser is designed to provide a better distribution of the cleaning pulse in
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN, AND O&M CONSIDERATIONS
2-16
-------
pulse-Jet fabric filters. This perforated tube, which fits inside the bag
cage, reportedly prevents the full force of the pulse from acting on the top
portion of the bag and allows more of the pulsed-air energy to reach the
bottom of the bag. The expected result is more uniform cleaning of the bag
19
and longer bag life.
The Electric Power Research Institute (EPRI) and the Public Service Co.
of Colorado (PSC) have tested the effectiveness of sonic energy enhancement
in removing accumulated dust from the bags of a reverse-air filter and keep-
on
ing the pressure drop down. In this method, horns sound during the re-
verse-air cleaning cycle to help dislodge the filter cake. Both EPRI and PSC
have had success with this technique; pressure drop has decreased as much as
on
40 to 50 percent. Further studies are being conducted by EPRI to determine
20
the impact of different horn frequencies, locations, timing, and duration.
All installations to date have been retrofitted. The effectiveness of this
technique in preventing excessive residual dust cake buildup on clean bags
has not yet been demonstrated.
2.3 SYSTEM COMPONENTS
2.3.1 Fabric
Filter fabrics commonly used in operating facilities are either felted
or woven. Felt is a genuine filter medium and collects particulates more
efficiently at comparable gas velocities; however, it is also more expensive.
Felted fabrics are composed of randomly oriented fibers and are relatively
thick. The thickness provides maximum particle impingement, but it also
increases the static pressure drop. Felted fabrics are normally used in
pulse-type units and are operated at high A/C ratios.
Woven fabrics are generally used in shaker and reverse-air filter sys-
tems and are operated at relatively low A/C ratios. Woven fabric is made up
of filament or staple (spun) yarns in a variety of weaves with various spac-
ings between the yarns. The finish is specifically designed to retain or
shed filter cake, depending on the application. Figure 2-7 shows seven of
the most common weave patterns. Plain weave has the lowest initial cost, the
least porosity, and the greatest particle retention; however, it also has the
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND OftM CONSIDERATIONS
-------
E
EJJ£P
Plain or Taffeta
Weave 1/1
3/2 Regular Twill
"Crowfoot" Satin or
4 Shaft Satin
No. 4 Harness
2/1 Regular
Twill
4/1 Satin
(Sateen if Cotton)
No. 5 Harness
3/1 Regular
Twill
a.
oc
o
K
O
2/2 "Broken" Twill
or Cham Weave
Other Popular Weaves:
Drill = 2/1 L.H. Twill, or 3/1 Twill
Herringbone - a type of broken twill
Basket Weave = extension of plain weave
Gabardine = regular or steep twill with
higher warp than fill count
Figure 2-7. Typical fabric weaves (reprinted by permission:
Industrial Gas Cleaning Institute).
2-18
-------
greatest potential for blinding. Twill weave has medium retention and blind-
ing characteristics and has reasonable permeability; it also exhibits the
best resistance to abrasion. Sateen weave has the lowest particle retention,
is easiest to clean of accumulated dust, and has the lowest potential for
blinding. The permeability of woven fabrics depends on the type of fiber,
the tightness of the twist, the size of yarn, the type of weave (geometric
pattern), the tightness of the weave (thread count), and the type of fabric
finish. The permeability of the fabric, however, has little to do with the
operation of the fabric under actual gas conditions.
Woven fabrics are available in several finishes. Cotton fabrics may be
preshrunk to maintain dimensional stability (i.e., resistance to stretching
or shrinking in any direction, which could adversely affect other fabric
characteristics). Spun fiber fabrics may be napped on the surface that will
receive the dust load. The napping process pulls fibers out of the yarn
bundles to form a soft pile; this promotes the formation of a dust cake on
the fabric surface that does not penetrate the interstices of the fabric.
Synthetic fabrics may be heat-set to ensure dimensional stability and to
provide a smooth surface with uniform permeability. Any fabric may be sili-
cone-treated (also used in combination with graphite and Teflon) to improve
abrasion resistance, to facilitate cake release, and to reduce moisture
absorption. Fabric finish is extremely important for inside bag filtering,
where nodules develop that restrict filtering (increase pressure drop) and
interfere with cake release. The formation of nodules can be reduced by
singeing the interior or exterior surface to remove fibers not tightly bound
in the weave.
Fabric selection is usually based on the prior experience in similar
applications. The following factors are important:
Dust penetration
Continuous and maximum operating temperatures
Chemical degradation
Abrasion resistance
Cake release
Pressure drop
Cost
Cleaning method
Fabric construction
Bag life desired
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O»M CONSIDERATIONS
-------
The choice of fabric ultimately affects pressure drop, selection of cleaning
method, outlet concentration, and the life of the fabric under operating
conditions.
Filter fabrics can further be broken down into the following three
classifications: natural, synthetic, and mineral. The fibers included in
each are as follows:
Natural
Cotton
Wool
Synthetic
Polypropylene
Nylon
Polyester (Dacron)
Nylon aramide (Nomex)
Acrylic (Dralon T)
Fluorocarbon (Teflon)
Polyphenylene sulfide (Ryton)
0 Mineral
Glass
Ceramic
Stainless steel (or other exotic alloys such as Haste!loy and
Chromalloy, depending upon the application)
Other substances can be used as coatings on these fabrics to improve
their operating characteristics in specific environments. The most common
coatings are Teflon and silicone/graphite.
The properties of commonly used fabrics and fabric coatings are dis-
cussed in conjunction with the gas stream properties that limit which fabric
can be used in various applications. These properties are temperature, the
chemical composition of the gas, and the abrasive nature of the particulate.
2.3.1.1 Gas Stream Properties Affecting Fabric Life—
Temperature—High temperatures accelerate the degradation of the polymer
in synthetic and natural fabrics. Table 2-1 lists the continuous tempera-
tures at which reasonable performance of the various fabrics can be expected
2-20
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY, DESIGN, AND O&M CONSIDERATIONS
-------
TABLE 2-1. RECOMMENDED TEMPERATURE LIMITS FOR VARIOUS FABRICS
1,2
Fiber
Cotton
Wool
Nylon
Dyne!
Polypro-
pylene
Orion
Dacron
Nomex
Teflon
Fiberglass
Stainless
Generic name
Natural fiber
cellulose
Natural fiber
protein
Nylon polyamide
Modacrylic
Polyolefin
Acrylic
Polyester
Nylon aromatic
Fluorocarbon
Glass
Maximum gas temperature, °F
Continuous
180
200
200
160
200
275
275
400
500
500
1000
Short-term
225
250
250
200
250
320
325
425
530
550
No data
Melting
temperature, °F
420 decomposes
575 chars
520
440 softens
440
520 softens
440
640 decomposes
670 decomposes
1070
1700
(Adapted from Perkins, H.C.
New York, 1974).
In: Air Pollution, A. Stern (ed.), McGraw Hill,
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY, DESIGN. AND O&M CONSIDERATIONS
2-21
-------
under normal conditions, according to fabric manufacturers. The maximum
short-term temperature represents the temperature at which rapid deteriora-
tion will occur and result in immediate failure. For synthetics, this is the
temperature at which polymer softening/polymer chain breakage occurs and
causes permanent elongation.
Polymer finishes, such as Teflon, can also degrade at high temperatures.
A few minutes of exposure to temperatures above the recommended continuous
operating temperatures may not result in immediate failure, but will reduce
the overall life of the fabric. The effects of repeated temperature excur-
sions on tensile strength are cumulative.
Chemical composition—Chemical degradation of the fabric is caused by
the breaking of polymer chains within the fiber structure. This can be
caused by acid hydrolysis or alkali attack.
As the chain length of a polymer is reduced by chemical attack, it loses
strength. The chemical attack may be accelerated by moisture or metal cata-
lysts in the dust impregnated in the fibers. The rate of attack increases
with temperature.
Chemical composition of the gas stream (along with its moisture content
and temperature) must be considered in selection of the fabric. Table 2-2
presents the ratings of commercial fabrics with respect to chemical resist-
ance. "Resistance" is a relative term; it does not imply total resistance to
a specific chemical. Also, resistance may be greatly reduced by cyclic oper-
ation under different conditions and concentrations. Tables such as this
must be used with caution. Polyester is generally rated as resistant to
alkali attack, but at temperatures above 93°C (200°F) and in the presence of
moisture, the polymer degrades rapidly. Cotton and Nomex are particularly
susceptible to sulfuric acid attack below the acid dew point. The tensile
strength of the fiber is reduced as the polymer chains are broken.
In general, the life of the fiber depends on proper fiber choice for
application to acid gases such as SCL, hydrogen chloride (HC1), and hydrogen
fluoride (HF).
Abrasive dust—How well a fabric resists abrasion depends on fabric
construction, fabric finish, and shapes of the particles collected. General
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND OftM CONSIDERATIONS
L.— L.L-
-------
TABLE 2-2. CHEMICAL RESISTANCE OF COMMON COMMERCIAL FABRICS1'2'15
Fiber
Cotton
Wool
Nylon
Dynel
Polypropylene
Orion
Dacron
Nomex
Teflon
Fiberglass
Polyethylene
Stainless steel
(type 304)
Generic name
Natural fiber
cellulose
Natural fiber
protein
Nylon polyamide
Modacrylic
Polyolefin
Acrylic
Polyester
Nylon aromatic
Fluorocarbon
Glass
Polyolefin
Acid resistance
Poor
Very good
Fair
Good to very
good
Excellent
Good to excel-
lent
Fair
Excellent
Fair to good
Very good to
excellent
Excellent
Fluoride
resistance
Poor
Poor to fair
Poor
Poor
Poor
Poor to fair
Poor to fair
Good
Fair to good
Poor
Poor to fair
Alkali-
resistance
Fair to good
Poor to fair
Very good to
excellent
Good to very
good
Excellent
Fair
Fair to good
Excellent
Excellent
Fair
Very good to
excellent
Excellent
Flex and
abrasion
resistance
Fair to good
Fair
Very good to
excellent
Fair to good
Very good to
excellent
Fair
Very good
Very good to
excellent
Fair
Poor
Good
ro
i
ro
GO
-------
abrasion of the fabric is accepted as a normal contributor to the failure of
a bag over its life. Failure due to abrasion cannot be prevented, but the
rate of abrasion can be reduced by properly installing the bags to avoid
bag-to-bag contact, by maintaining proper tension, and by reducing the amount
of dust being handled. In some applications, a cyclone may be installed as a
precleaner to remove larger particles and reduce inlet loading. In other
applications, however, the removal of larger particles may not be desirable
because of dust cake considerations.
2.3.1.2 Other Factors Affecting Bag Life--
Local intensive abrasion—Local intensive abrasion, which can cause
premature bag failure, can be prevented. High abrasion rates are commonly
associated with improper bag installation or design flaws in the collector.
Higher A/C ratios will also increase abrasion. Each case of abrasion failure
must be addressed separately to determine if corrective action may be taken
to reduce failure frequency.
Cleaning method--The method of removing the dust cake is closely related
to fabric construction and fiber type. With woven fabrics that are subject
to abrasion or flex damage, gentle cleaning methods such as low-frequency
shaking or reverse air can be used. Felted fabrics require a more intense
cleaning method, such as high-pressure reverse-air or pulse-jet cleaning.
Use of an improper cleaning method with a fabric (e.g., intense shaking of
glass bags) can cause premature fabric failure, incomplete cleaning, or
blinding of the fabric (complete plugging of pores).
Pressure drop across filter—Greater stress on the fabric at higher
pressure drops can decrease bag life.
2.3.2 Instrumentation
The following instrumentation contributes to the reliable operation of a
fabric filter.
1. Thermocouples or other temperature-measuring instruments at the
device inlet.
2. Inlet/outlet differential static pressure gages.
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND OftM CONSIDERATIONS
-------
3. A double-pass transmissometer (opacity meter).
4. Compressed-air pressure gage.
5. Fan motor ammeter.
Recording temperature meters are especially useful in identifying high-
er low-temperature excursions, which rapidly destroy fabrics. High-tempera-
ture indicators composed of colored fiber or temperature-sensitive plugs may
be used as a less expensive alternative.
In lieu of differential pressure gages, it is sometimes simpler to
install static pressure taps, where appropriate, and to use a portable meter
to obtain readings. This approach reduces problems of meter moisture damage,
meter corrosion, and plugging of lines. If permanent differential static
pressure gages are used, the static pressure lines should be as short as pos-
sible and free of 90-degree elbows to minimize plugging. Copper tubing in a
noncorrosive environment has been found to be less susceptible to deteriora-
tion than the polypropylene lines commonly used. However, PVC is even better
because of its resistance to corrosive conditions.
The double-pass transmissometer may not provide an accurate measurement
of effluent opacity; however, it is useful in identifying problems. A sig-
nificant leak is detected in a specific compartment by a drop in the opacity
when that compartment is off-line for maintenance in shaker and reverse-air
21
fabric filters. A pinhole leak can be identified in pulse-jet filters by
an increase in opacity a few seconds after the row of bags that contains the
pinhole leak is cleaned.
The instrument readouts should be mounted on a master control panel as
close as possible to other process monitoring displays. The readings of
thermocouples, pressure differential gages, and transmissometers can all be
electronically recorded for permanent records.
2.3.3 Gas inlet equipment
If an adequately designed baffle plate is not used to remove large,
abrasive particles, abrasion can occur, particularly on the surface of the
bag near the cuff. More abrasion problems occur near the bottom of the bags,
directly above the thimbles. The installation of thimble extensions, a blast
plate, or a precleaner can reduce the abrasive damage and extend bag life.
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND OftM CONSIDERATIONS
2-25
-------
Baffle plates or diffusers are used to cause large particles to deposit
in the hopper before they contact the bags. The orientation of the plates is
critical, however, because deflection of incoming gas into the hopper can
resuspend collected dust and increase dust loading through the tube sheet.
Less resuspension will occur if the hoppers are operated with continuous dust
removal and the dust remains below the gas inlet. Screw conveyor discharge
also should be on the opposite side from the gas inlet to minimize reentrain-
ment. Because baffle plates suffer continuous erosion, they must be replaced
periodically. An increase in bag ruptures near the cuff area may indicate
the need to replace a baffle plate or to correct other problems with the tube
sheet thimble.
A good thimble arrangement will also reduce abrasion at the bottom of
bags in reverse-air and shaker-type collectors. The thimbles should be at
least one bag diameter long to prevent abrasion caused by particulate "turn-
ing the corner" at the cell plate and being thrown to the outside by inertial
22
force. The thimbles act as flow straighteners and protect the bottom of
the bag from excessive abrasion. A properly designed unit is shown in Figure
2-8. The rounded edge on the top of the thimble reduces cutting of the
fabric even if tension is not optimum. For units in which the base snaps
into the tube sheet, a thimble can be added that extends downward to provide
the same type of abrasion protection as that shown in the illustration.
A bypass may be advisable, especially when process startup or upset con-
ditions could generate sticky particulate or result in gas temperatures below
the acid vapor or water dew points. These could also be used in conjunction
with a spark sensor to reduce risk of fire.
Inlet and outlet dampers should be provided in compartmented systems to
allow on-line maintenance. The dampers must be designed to provide positive
sealing so as to protect maintenance personnel from toxic gases.
2.3.4 Hoppers and Dust Handling System
As solids are cleaned from the filtering fabric, they fall into a
collection hopper for ultimate removal. The "fluid" properties of the
collected solids are important to the design and operation of these systems,
and they may be markedly different from the properties of the material from
which they originated. Fine dusts, for example, tend to pack more readily
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN, AND O&M CONSIDERATIONS
-------
«* BAG
CLAMP *~Q
D
THIMBLE
I— TUBE SHEET
Figure 2-8. Cross section of a thimble protecting bottom of bag
(Courtesy of Mr. E. W. Stanly).
2-27
-------
than coarser materials; moreover, condensation formed in the filter device
may cause solid material to agglomerate. Beth of these factors can make
solids disposal difficult.
Various design features can help to prevent the clogging of solid col-
lection hoppers. The hopper should be designed with a steep valley angle;
angles of 55 to 70 degrees are recommended. Hoppers should also include
large discharge openings, well finished (smooth) surfaces, and minimal ledges
or other obstructions on sidewalls. The top of the hopper sidewall should
drop vertically and the slope to the discharge point should begin at least
one bag diameter below the bottom of the bags to allow proper dust discharge.
At least 1 ft of clearance should be provided between hopper walls and any
internal partitions to allow easy discharge.
Heaters and insulation can be installed in hoppers to prevent condensa-
tion and caking of collected material. Hopper stones supplied with hot dry
air can also be used to fluidize material in the hopper and keep it free-
flowing.
Solids are generally removed from the hopper by means of a discharge
valve, which removes ash from the hopper while preserving the pressure dif-
ferential between the dust conveyance system and the fabric filter system.
Insulating the fabric filter helps to maintain the temperature of the
gas while it is being cleaned. Temperatures above the acid dew point mini-
mize corrosion of hanger components, doors, and walls. Internal shell compo-
nents can also be lined with appropriate corrosion-resistant materials if
necessary. Insulation should be applied to the baghouse shell, hoppers, and
doors. Structural steel may be placed on the internal portion of the shell.
The insulation may then be applied evenly across the shell to reduce "cold
spots," which promote corrosion of the shell.
2.3.5 Bag Support System
The bags and the associated fabric filter components (such as the tube
sheet, clamps, thimbles, and bag cages) must be compatible for optimum bag
life and control efficiency.
In shaker and reverse-air fabric filters, the bag can be attached to the
tube sheet by a thimble and clamp-ring design or by a snap-ring design. Fig-
ure 2-9 shows the two methods of attachment. Dust enters the fabric filter
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O&M CONSIDERATIONS
L.-C.O
-------
THIMBLE AND CLAMP RING DESIGN
POOR
CLAMP
BAG
SHORT
THIMBLE
,INCREASED
«r ABRASION
SHORT CUFF
J
TUBE SHEET
7
GAS FLOW
BETTER
BAG
LONG
f THIMBLE t
GAS FLOW
LONG CUFF AND
REDUCED ABRASION
TUBE SHEET
POOR
BAG
CUFF
WITH
SNAP
RIME
POOR
SNAP RING DESIGN
SHORT CUFF
NO THIMBLE
INCREASED
^ABRASION
'BETTER
BAG
CUFF WITH
SNAP RING
V6AS I
LONG CUFF AND
REDUCED ABRASION
TUBE SHEET
AND
THIMBLE
6AS FLOW
Figure 2-9. Methods of bag attachment in shaker and reverse-air fabric filters.
(Courtesy of PEI Associates, Inc.)
2-29
-------
at the hopper in a horizontal direction and must make a vertical turn to
enter the tube sheet thimbles. Because heavy particles with higher inertia
do not follow the flow, they do not enter the opening parallel to the thimble
walls. The particles impact on the walls of the thimble and, if the thimble
is short, on the fabric above the thimble. The action of the particles
striking at an angle to the fiber surface increases abrasion. Roughly 90
percent of bag failures occur near the thimble. The use of double-layered
fabric (cuffs) or longer thimbles reduces the failure rate.
In the snap-ring system, no thimble is used, and in some cases, a cuff
is not used. This exposes the bag to rapid abrasion a few inches above the
snap ring. Add-on tube sheet thimbles may be used to reduce this abrasion.
In the no-thimble design, improper installation of the snap ring can
result in dust penetration between the tube sheet and the bag cuff (see
Figure 2-10). If the ring seating is questionable, the bag should be removed
and reinstalled. If an adequate fit cannot be achieved, the bag should be
discarded.
Proper bag tension is important to ensuring adequate bag life and mini-
22 2°
mum particulate emissions. ' ~ Figure 2-11 shows a typical spring bag ten-
sioning mechanism. The bag should be tight enough to provide optimum utili-
zation of cleaning energy. It may be necessary to check bag tension soon
2"?
after startup. For uniform cleaning efficiency, the tension must be uni-
form in all the bags. Proper tension also reduces bag failures at the cuff,
lessens wear on thimbles, and improves cleaning efficiency. It should also
be noted that tension varies throughout the cleaning cycle and with bag age.
2.4 FABRIC FILTER O&M CONSIDERATIONS
This subsection sets the stage for succeeding manual sections, which ad-
dress definitive O&M procedures, methods, and practices that promote reliable
system performance and integrity. Theoretically, fabric filters can achieve
mass collection efficiencies in excess of 99.5 percent when particles are as
small as 0.01 ym. In practice, however, many process conditions and instal-
lation problems can reduce both the collection efficiency and the time avail-
able for service. Fabric filters require proper operation, extensive preven-
tive maintenance, and periodic inspections aimed toward reducing periods of
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O&M CONSIDERATIONS o
-------
SNAP RING WOUND
WITH FIBER
Figure 2-10. Proper method of installing bag in tube sheet with snap rings.
(Courtesy of PEI Associates, Inc.)
2-31
-------
J-BOLT
LOCK NUT
WASHER
SPRING
>
FRAME
WEDGE SEAL CAP
BAG
BOLT
TENSIONING
MECHANISM
Figure 2-11. Typical spring bag tensioning arrangement,
(Courtesy of Industrial Clean Air)
2-32
-------
excess emissions. Each of these is discussed in detail in the remaining
sections of this manual, but they are touched on briefly here.
2.4.1 Causes of Poor Performance
Giving proper attention to the design, construction, and installation of
the fabric filter system will enhance an O&M program. The design and con-
figuration should allow accessibility to all operating components and the
interior of the fabric filter for performance monitoring (Section 3), for
troubleshooting (Section 4), for performing preventive maintenance (Section
5), and for making detailed inspections (Section 6). On larger conpartmented
fabric filter systems, each compartment should be dampered (both automatical-
ly and manually) for isolation so that the entire fabric filter does not have
to be off-line to perform needed maintenance. The installation of a fabric
filter should be closely monitored by in-house environmental personnel for
quality assurance during construction.
Of particular importance are welding of the tube sheet and housing, in-
sulation work, and bag installation. Improper insulation, for example, can
lead to condensation inside the fabric filter, and deterioration of the bags
can result from chemical attack or bag blinding. Such deterioration will
lead to premature bag failures, greater emissions, and increased maintenance
costs for bag replacement. Leaks due to tube sheets not being completely
welded to the filter housing and bags not being properly installed allow
unfiltered exhaust gases to bypass the filter material (see Figures 2-12 and
2-13). Inferior workmanship and materials may lower the capital costs, but
efficiency of the unit will suffer.
Troubleshooting is an important part of operation and maintenance. All
potential causes of a problem should be investigated, not just the most obvi-
ous. Fabric failure, for example, is an obvious cause for a loss in particu-
late removal efficiency. Merely replacing the bag, however, may only tempo-
rarily solve the problem. Problems with the cleaning system, solids removal
system, or other factors may have contributed to bag deterioration. In other
words, just treating the symptoms does not always effect a cure. Section 4
details the techniques for effective performance evaluation, problem diag-
nosis, and correction.
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O»M CONSIDERATIONS
w "" 0*5
-------
Figure 2-12. Tubesheet leak due to poor welding at wall of pulse-jet
fabric filter. This results in a normal pressure drop but high continuous opacity.
(Courtesy of PEI Associates, Inc.)
IX)
CO
-------
Figure 2-13. Broken weld on roof of fabric filter. This allows
significant quantities of water to enter, and the water-sprayed bags
can become blinded.
(Courtesy of PEI Associates, Inc.)
2-35
-------
A good routine preventive maintenance program (Section 5) can minimize
typical performance problems such as bag blinding, cleaning mechanism fail-
ure, and pinhole leaks (Section 3). Many of these problems, for example, can
be minimized by routine maintenance of insulation, motors associated with the
cleaning mechanisms and dust removal system, and temperature control instru-
ments .
2.4.2 Establishing an Adequate Operation and Maintenance Program
Why should a plant make a concerted effort to maintain its fabric filter
properly? The most convincing reason, outside of the necessity to meet
applicable particulate emission regulations, is one of economics. A fabric
filter is an expensive piece of equipment, and even well-designed equipment
will deteriorate rapidly if improperly maintained and will have to be re-
placed long before it should be necessary. Not only can proper O&M save the
plant money, it can also contribute to good relations with the local control
agency by showing good faith in its efforts to comply with air regulations.
A fabric filter is unlikely to receive proper O&M without management
support and the willingness to provide its employees with proper training.
The fabric filter must be elevated to the same level of importance as the
process, and the operator must know the difference between a process change
and deterioration of the fabric filter. Management must instill an attitude
of alert, intelligent attention to the operation of the fabric filter instead
of waiting for a malfunction to occur before acting. This requires a con-
sistent monitoring program entailing the maintenance of detailed documenta-
tion of all fabric filter operations.
Although each plant has its own method of conducting an O&M program,
past experience has shown that plants that assign one individual the respon-
sibility of tying all the pieces of the program together operate better than
those where different departments look after only a certain portion of the
program and have little knowledge of how that portion impacts the overall
program. In other words, a plant needs to coordinate the operation, mainte-
nance, and troubleshooting components of its program if it expects to be on
top of the situation.
Some companies that have several plants have found it to be advantageous
to set up a central coordinating office to monitor the O&M status at each
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O&M CONSIDERATIONS
2-36
-------
plant. The resulting improved communications can provide an opportunity to
develop standardized reporting forms, assistance in personnel training,
interpretation of operating data, and routine inspections. With knowledgeable
people in the central coordinating office, the plants have somewhere to go
for assistance in solving problems for their specific kind of fabric filter.
Another resource that plants can draw upon is the manufacturer's field
service engineer. This person is involved in pre-operational inspections to
ensure proper assembly of fabric filter components; to set up the various
controls within prescribed limits; to check for proper operation of the dust
discharging system; to solve any existing problems after initial startup; and
finally, to instruct plant personnel on how to perform these functions.
Experienced field service engineers can be very helpful as a resource
for assistance in troubleshooting; however, because manufacturers are gener-
ally plagued with a high turnover rate, the plant should be wary of inexperi-
enced people who may incorrectly diagnose operating problems or be unaware of
proper correction procedures. This only adds to the confusion by misleading
O&M personnel.
The training and motivation of employees assigned to monitor and main-
tain the fabric filter are critical factors. These duties should not be
assigned to inexperienced people who do not understand how the fabric filter
works or the purpose behind their assigned tasks. The employee must know
what management expects and should receive encouragement for a job well done.
Regular training courses should be held by in-house personnel or by the
use of outside expertise so that operators and maintenance personnel are
instructed on everything they need to know in regard to the fabric filter.
This should include written instructions and "hands-on" sessions on safety,
how to make inspections while the fabric filter is both in and out of serv-
ice, how to take operating parameter readings, how to perform routine mainte-
nance, and how to record and use data. Training provides the knowledge
necessary for proper operation and maintenance of the fabric filter and makes
the employees' job easier because they will understand why they are taking
pressure and temperature readings or searching for pinhole leaks.
In summary, the three separate components of an adequate plan for long
fabric filter life are operation, maintenance, and troubleshooting. Each
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O&M CONSIDERATIONS
c. ~*5 /
-------
plant should have its own O&M procedures manuals, blueprints, and a complete
set of fabric filter specifications; an adequate supply and record of spare
parts; written procedures for addressing malfunctions; and formalized audit
procedures.
Records should be kept on fabric filter operating conditions (process
logs, fuel records, gas temperature, pressure differentials, etc.)» equipment
conditions (internal inspections; daily inspections of cleaning mechanisms,
hoppers, compressors, etc.), maintenance (work orders, current work in
progress, deferred work), and troubleshooting/diagnostic analysis (component
failure frequency and locations, impact of process changes on fabric filter
performance, and other trend-related analyses). Each of these areas is
discussed in detail in later sections of this manual.
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN, AND OftM CONSIDERATIONS
-------
REFERENCES
1. Theodore, L., and A. J. Buonicore. Industrial Air Pollution Control
Equipment for Particulates. CRC Press, Cleveland. 1976.
2. Perkins, H. C. In: Air Pollution, A. Stern (ed.). McGraw-Hill, New
York. 1974.
3. Paretsky, L. L., T. R. Pfeffer, and A. J. Squires. J. Air Pollution
Control Association, 21:4, April 1971.
4. Anderson, D. M., and L. Silverman. Harvard Air Cleaning Laboratory.
AEC Report No. NYO-4615, 1958.
5. Donavan, R. P., J. H. Turner, and J. H. Abbott. Passive Electrostatic
Effects in Fabric Filtration. Second Symposium on the Transfer and
Utilization of Particulate Control Technology. In: Vol. 1. Control of
Emissions from Coal Fired Boilers. F. P. Venditti, 0. A. Armstrong, and
M. Durham (ed.). EPA-600/9-80-039a, September 1980. pp. 476-493.
6. Donavan, R. P., R. L. Ogan, and J. H. Turner. The Influence of Electro-
statically Induced Cage Voltage Upon Bag Collection Efficiency During
the Pulse-jet Fabric Filtration of Room Temperature Fly Ash. In: Pro-
ceedings of the Third Symposium on Fabric Filters for Particle Collec-
tion. EPA-600/7-78-087, June 1978.
7. Frederick, E. J. of the Air Pollution Control Association, 24:1164-
1168, December 1974.
8. Frederick, E. Chemical Engineering, 68:107, June 1971.
9. Dennis, R., and N. F. Suprenant. Research on Fabric Filtration
Technology. EPA-600/8-78-005d. 1978. p. 151.
10. Dennis, R., and N. F. Surprenant. Research of Fiber Filtration Technol-
ogy. EPA-600/8-78-005d, 1978, p. 151.
11. Bergmann, L. New Fabrics and Their Potential Application. J. Air
Pollution Control Association, 24:1187-92, December 1974.
12. Mappes, T. E., and R. D. Terns. An Investigation of Corrosion in Partic-
ulate Control Equipment. EPA-340/1-81-002, February 1981.
2-39
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN, AND O&M CONSIDERATIONS
-------
13. Prinz, R. T. Reducing Baghouse Maintenance by Design. Minerals
Processing, 11:8-13, May 1970.
14. Dennis, R., and H. A. Klemm. Fabric Filter Model. GCA Corporation,
Bedford, Massachusetts. EPA-600/7-79-043a through EPA-600/7-79-043c,
February 1979.
15. Reigel, S. A. Fabric Filtration Systems Design, Operation, and Mainte-
nance. From Plant Inspection Workshop - Techniques for Evaluating
Performance of Air Pollution Control Equipment. Selected Papers on
Operation and Maintenance of Fabric Filters. Compiled by PEDCo Environ-
mental, Inc., Durham, N.C., under U.S. Environmental Protection Agency,
Contract No. 68-02-3512. February 1981. pp. 35-73.
16. Wheelabrator-Frye, Inc. Clearing the Air With Custom-Designed Fabric
Filter Systems. Pittsburgh. Sales Literature. APC-11 15-7-7. 24 pp.
17. Pruce, L. M. Interest in Baghouses on Upswing. Power, February 1980.
p. 87.
18. Burnett, T. A., and K. D. Anderson. Technical Review of Dry FGD Systems
and Economic Evaluation of Spray Dryer FGD Systems. EPA-600/7-81-014,
February 1981.
19. Lucas, R. L., et al. The Staclean Diffuser Increases Capacity and
Reduces Bag Wear in Pulse-Jet Baghouses. Presented by duPont de Nemours
& Company at the 73rd Annual Meeting of the Air Pollution Control
Association, Montreal, Quebec, June 22-27, 1980. 16 pp.
20. Carr, R. C. Summary of Second Conference on Fabric Filter Technology
for Coal-Fired Power Plants. J. of the Air Pollution Control Associa-
tion, 33(10)949-954, October 1983.
21. Perkins, R. P. The Case for Fabric Filters on Boilers. Presented at
Semiannual Technical Conference on Air Pollution Equipment, Philadel-
phia, Pennsylvania, April 23, 1976.
22. Ladd, K. L., Jr., et al. Objectives and Status of Fabric Filter Per-
formance Study. In: Second Symposium on the Transfer and Utilization
of Particulate Control Technology. Vol. 1, Control of Emission from
Coal-Fired Boilers. F. P. Venditti, F. A. Armstrong, and M. Durham
(ed.). EPA-600/9- 0-039a, September 1980. pp. 317-341.
23. Perkins, R. P., and J. F. Imbalzano. Factors Affecting Bag Life Per-
formance in Coal-Fired Boilers. In: Proceedings of The User and Fabric
Filtration Equipment III. E. R. Fredrick (ed.). APCA Specialty Confer-
ence. 1978. pp. 120-144.
?4. PEDCo Environmental, Inc. Inspection and Operating and Maintenance
Guidelines for Secondary Lead Smelters. Prepared for U.S. Environmental
Protection Agency, Cincinnati, Ohio. June 1983. p. 66.
SECTION 2-OVERVIEW OF FABRIC FILTER THEORY. DESIGN. AND O&M CONSIDERATIONS
2-40
-------
SECTION 3
FABRIC FILTER PERFORMANCE MONITORING
Performance monitoring is a key factor in establishing good operation
and maintenance procedures for a fabric filter. It includes measurement of
key operating parameters by both continuous and intermittent methods, compar-
ison of these parameters with baseline and/or design values, and the estab-
lishment of recordkeeping practices. These monitoring data are useful in
performance evaluation and problem diagnosis. In this section, the key
operating data and procedures used in performance monitoring are discussed.
Interpretation of the data is covered in Section 4.
3.1 KEY OPERATING PARAMETERS AND THEIR MEASUREMENT
Several operating parameters are indicative of a likely change in per-
formance. Some of these parameters are easily measured and monitored on a
continuous basis, whereas others must be measured only periodically because
of the expense and/or difficulty in measurement. Most of these parameters,
however, directly affect fabric filter performance. The following typical
parameters are discussed here: gas volume and gas velocity through the
fabric filter; temperature, moisture, and chemical composition of the gas;
particle size distribution and concentration; and pressure drop across the
fabric filter. Many of these factors are interrelated and affect critical
fabric filter performance factors such as air-to-cloth ratio, the required
cleaning energy and its effectiveness, and bag tension.
3.1.1 Gas Volume and Velocity
A change in gas volume and velocity affects the air-to-cloth ratio, the
required cleaning energy and its effectiveness, and bag life. The collection
efficiency also varies with fabric filter designs, but such variations are
generally not significant unless the flow is increased beyond the design
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
-------
limitations. Higher gas volumes lead to higher air-to-cloth ratios and
velocities, which shorten bag life as a result of more frequent cleaning,
higher particle velocity through the fabric (more abrasion), greater poten-
tial for blinding, and a higher pressure drop.
A pitot tube traverse is normally used to measure total gas volume, and
the method is usually a combination of EPA Reference Methods 1 and 2. In
this method, the duct on the cross section of the stack is divided into a
number of equal areas, and each area is sampled to arrive at an average
velocity through the duct. When the average velocity and the duct or stack
cross-sectional area is known, the average gas volume can be determined.
Because most facilities do not routinely measure gas volume, other indirect
indicators may be used to estimate the volume. These include fan operating
parameters, production rate, and a combination of other gas condition parame-
ters.
3.1.2 Gas Temperature
Monitoring the temperature of the gas stream can provide information
about the performance of a fabric filter and clues for diagnosing both fabric
filter performance and process operating conditions. The major concern in
temperature measurement is to avoid sampling at a stratified point where the
measured temperature is not representative of the bulk gas flow. Thermocou-
ples with digital, analog, or strip-chart display are typically used.
The effect of temperature is most important as an indicator of excessive
inleakage into the gas stream. Even the best-constructed and best-insulated
fabric filter will experience some temperature drop, which can be as low as
1° to 2°F on smaller fabric filters and up to 25°F on very large fabric fil-
ters. Of course, the expected temperature drop will vary, depending on the
temperature of the gas stream relative to ambient temperatures. In fact, the
temperature drop is normally higher on small collectors than on large ones
because the ratio of the outside collector area to gas throughput is general-
ly higher. In any case, some acceptable or maximum difference between inlet
and outlet gas temperatures should be set, which when exceeded, would indi-
cate improper operation or a maintenance problem that requires correction.
Temperature monitoring is also important for minimizing bag damage due
to temperature exposure above the fabric's design limits. Even a temporary
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-2
-------
excursion above fabric temperature limits can weaken bags. An alarm tied to
a regularly maintained thermocouple probe may prevent bag failure due to tem-
perature excursion. This alarm can be used in conjunction with an automatic
method of protection, such as dilution air or bypass. The effects of re-
peated temperature excursions are cumulative, and temperature charts can be
used to determine the potential for short-term failure due to temperature
excursions. It should be noted that the use of only one probe does not
represent the temperature in each section of the fabric filter, but rather
only the average.
Both maximum and minimum fabric filter operating temperatures must be
considered. The exposure of the fabric to temperatures above the maximum
exposure temperatures can cause immediate failure due to complete loss of
strength and permanent elongation (melting). Minimum temperatures are relat-
ed to the dewpoint temperature of the gas stream. Operation of the fabric
filter below this temperature can cause moisture or acid condensation and
result in bag blinding or chemical attack of the fabric. Fabric life under
these conditions depends on the proper initial choice for application to acid
gases such as SCL, hydrogen chloride (HC1), and hydrogen fluoride (HF).
3.1.3 Chemical Composition
Important factors regarding the chemical composition of the gas stream
include moisture content and acid dewpoint. Operating a fabric filter at
close to the acid dewpoint presents a substantial risk of corrosion, espe-
cially in localized spots close to hatches, in dead air pockets, in hoppers,
or in adjacent heat sinks (such as external supports). If the operating
temperature drops below the water dewpoint, either during startup or under
normal operation, blinding of the bags can occur. Trace components such as
fluorine also can attack certain fabrics. For example, fiberglass bags
exposed to 200 ppm of HF at 500°F may last 2 years, while 1200 ppm of HF at
500°F may result in a bag life of only 2 months.
From a practical standpoint, the chemical composition of the dust and
gas stream is a dynamic quantity, and any monitoring scheme can only point
out an optimum range and the variability. Monitoring the level of certain
compounds may prove useful in some instances; for example, in the combustion
of coal, sulfur content, combustibles content, and chemical composition of
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-3
-------
the ash may provide supporting evidence when problems occur. In many in-
stances, however, chemical composition is either not monitored or it is
monitored for other purposes.
3.1.4 Mass Loading and Size Distribution
Mass loading and size distribution must be considered during the design
of a fabric filter and also during operation; however, within certain limita-
tions (±10 to 20 percent of design values), changes in these parameters do
not seriously affect fabric filter efficiency. Nevertheless, an increase in
mass loading may require more frequent cleaning of the bags as a result of
faster filter cake buildup. When bag failures occur due to the presence of
large abrasive particles, the use of a precleaner or a gas distribution
device (i.e., an inlet diffuser) at the fabric filter inlet may be required.
For some sources, such as spreader stoker boilers, installation of a mechani-
cal collector ahead of the fabric filter may be necessary to protect the bags
from the large number of >10-micrometer particles and from glowing embers.
Mass loading at the inlet and outlet of the fabric filter is usually
measured by standard EPA reference methods. The difference between the
amount of material in the outlet gas stream and the inlet gas stream provides
the basis for removal efficiency calculations. The use of the reference sam-
pling methods, however, can be difficult on processes that generate very high
mass loadings at the fabric filter inlet. When outlet mass loadings are very
low, long sampling times may be required to collect enough material for accu-
rate weighing. Also, simultaneous sampling of inlet loadings during the en-
tire test period may not always be possible if the loadings are so high that
the sampling train becomes overloaded. In some instances, a series of probes
inserted for 1 to 15 minutes to take "grab" samples of the inlet concentra-
tion may be all that is technically feasible. Although this may not provide
as accurate a value for inlet mass loading as would an "integrated" sample
taken concurrently with the outlet emissions test, it will give a reasonable
value to work with.
Particle size distribution is usually determined through the use of
cascade impactors. Various types of cascade impactors are available with
different particle cut sizes and for different mass loadings. A typical
cascade impactor system is presented in Figure 3-1. The cascade impactor is
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-4
-------
JET SIZE
JET VELOCITY @
.0100" Dia.
154 FT/SEC
.O100" Dia.
77.0 FT/SEC
.0135" Dia.
42.3 FT/SEC
.021O" Dia.
17.50 FT/SEC
.0280" Dia
9.81 FT/SEC
.036O" Dia.
5.91 FT/SEC
.0465" Dia.
3.57 FT/SEC
.0636" Dia
1.91 FT/SEC
GASKET
(TYP)
Figure 3-1. Typical cascade impactor system (courtesy of Andersen Samplers, Inc.)
3-5
-------
usually placed on a standard sampling probe and inserted into the gas stream
for isokinetic sampling of the particulate. A sampling train with a cascade
impactor is illustrated in Figure 3-2. After sampling is completed, each
stage of the impactor is weighed in the lab and compared against its initial
weight to determine distribution. Because the impactor consists of several
stages (usually five to nine) and each stage corresponds to a progressively
smaller particle size range, the weight gain of each successive stage pro-
vides a weight distribution of particle sizes.
Cascade impactors have two limitations: the flow rate cannot be varied
during the test run, and multiple-point samples are not usually possible with
a single sample train. The sampling location must be selected carefully to
avoid stratification and to provide a representative sample that will produce
valid results. The particle capture characteristics of a cascade impactor
are calibrated against a given flow rate. Thus, the stated particle size
range for any given stage in the impactor is referenced against a fixed flow
rate. Changes in the reference flow rate to provide isokinetic sampling in
the stack will change the particle size range that each impactor stage will
capture. If the chosen flow rate is different from the reference value,
calibration curves are available for each impactor to correct for changes in
the particle size sensitivity of each impactor stage. Thus, the flow rate
through the impactor cannot be changed once it has been established. This
necessitates single-point sampling, which is essentially a grab sample. The
situation is even worse at the inlet, where sample times may be limited to
only 1 to 2 minutes because of mass loading. More than a single-point sample
may be obtained by the use of multiple cascade impactors to sample a number
of different points. This is both equipment- and labor-intensive; however,
it may provide an indication of the representative nature of a single-point
sample.
3.1.5 Pressure Drop
Each fabric filter is designed to operate at a specific pressure drop or
within a certain range. If the fabric filter is operating properly, the
pressure drop should remain fairly steady during normal operation, with a
gradual increase as the filter cake builds up on the bags and a steep de-
crease immediately after the bags are cleaned. Pressure measurements alone
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-6
-------
METER BOX
PROBE TUBE
Figure 3-2. Sampling train with cascade impactor.
(Courtesy of PEI Associates, Inc.)
3-7
-------
indicate the permeability of the cloth, how heavy the dust deposit is before
cleaning, how complete the cleaning is, and whether the fabric is starting to
plug or blind. It should be noted, however, that since AP is a function of
velocity, values can only be compared at the same volume flow to show fabric
change.
Static pressure gauges (e.g., a magnehelic gauge or manometer) should be
installed at the inlet and outlet of the fabric filter to determine the
pressure drop across the unit. In many applications, static pressure indica-
tors must withstand high temperature and dust loadings. Because the most
common problem with pressure indicators is plugging of the taps, provisions
for cleaning the taps must be included. At facilities that use pulse-jet
fabric filters for fugitive emission control, a pressure indicator should be
installed to determine the pulse header pressure. An alarm also can be
connected to such an indicator to signal the operator when pulse pressure
o
drops below a preset value (e.g., 70 Ib/in. ).
It is sometimes simpler to install static pressure taps in lieu of
differential pressure gauges, where appropriate, and to use a portable meter
to take readings. This approach reduces problems of meter moisture damage,
meter corrosion, and plugging of lines. When permanent differential static
pressure gauges are used, the static pressure lines should be as short as
possible and free of 90-degree elbows. As mentioned in Section 2, copper
tubing (in a noncorrosive environment) has been found to be less susceptible
to deterioration than the polypropylene lines commonly used. However, PVC is
even better because of its resistance to corrosive conditions.
3.1.6 Bag Tension
Proper bag tension is an important factor for ensuring adequate bag life
and minimum particulate emissions. The bag should be tight enough to avoid
excessive fiber-to-fiber and bag-to-bag abrasion, but not so tight as to
exceed the tensile strength of the bag during cleaning. An example of a
properly tensioned bag is shown in Figure 3-3.
Upon installation, each bag must be checked for proper tension. The
manufacturer's literature should be consulted to determine the correct ten-
sioning method. The proper bag cleaning requires flexing of the surface to
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-8
-------
Figure 3-3. Example of a properly tensioned bag that is
collapsed during reverse-air cleaning.
(Courtesy of PEI Associates, Inc.)
3-9
-------
dislodge the cake, and if bag tension is low, the bag may be flexed adequate-
ly at the top, but the standing wave will dampen as it moves downward. Also,
the fabric may elongate because of the weight of dust collected between
cleaning cycles, or the bag attachments may slip in the hangers. Thus,
tension may change with length of service, and it also should be checked soon
after startup and periodically thereafter.
3.2 INSTRUMENTATION SYSTEMS AND COMPONENTS
Numerous instruments may be used to monitor fabric filter performance
and performance changes. These include the usual pressure and temperature
sensors, transmissometers, and hopper level indicators.
3.2.1 Transmissometers
Transmissometers can be useful for determination of fabric filter per-
formance levels. A facility may have one or more monitors that indicate
opacity from various fabric filter outlet ducts and from the stack itself.
Opacity also may be measured on a real-time basis or over selected averaging
periods.
The opacity monitor simply compares the amount of light generated and
transmitted by the instrument against the quantity received by the receiver.
The difference, which is caused by absorption, reflection, refraction, and
light scattering by the particles in the gas stream, is the opacity of the
gas stream. Opacity is a function of particle size, concentration, and path
length. Opacity monitors are typically calibrated to display opacity at the
stack outlet path length. Most of the opacity monitors now being installed
are double-pass monitors; i.e., the light beam is passed through the gas
stream and reflected back across to a transceiver. This arrangement is
advantageous for several reasons: 1) it allows automatic checking of the
zero and span of the monitor when the process is operational; 2) because the
path length is longer, the monitor is more sensitive to slight variations in
opacity; and 3) all of the electronics package is located on one side of the
stack as a transceiver. Although single-pass transmissometers are available
at a lower cost (and sensitivity), the double-pass monitor can meet the
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
-------
requirements for zero and span in Performance Specification 1, Appendix B, 40
CFR 60. Monitor siting requirements are also discussed in Performance Speci-
fication 1.
For many sources, mass-opacity correlations can be developed to provide
a relative indication of fabric filter performance. Although site-specific,
these correlations can provide plant and agency personnel with an indication
of relative performance levels at a given opacity and deterioration in per-
formance that requires attention by plant personnel.
When parallel fabric filters or chambers are used, an opacity monitor
can be placed in each outlet duct, as well as on the stack, to measure the
opacity of the combined emissions. Although the stack monitor is commonly
used to indicate stack opacity (averaging opacities from different ducts can
be difficult), the individual duct monitors can be used to determine bag
integrity in each chamber and for troubleshooting. Although this option is
often not required and it represents an additional expense, it can be very
useful, particularly on relatively large fabric filters.
3.2.2 Hopper Level Indicators
Hopper level indicators could more accurately be called high hopper lev-
el alarms because they do not actually measure dust levels inside the hopper;
instead, when the dust level becomes higher than the level detector, an alarm
sounds to indicate that corrective action is necessary. The level detector
should be placed high enough that "normal" dust levels will not continuously
set off the alarm, but low enough to allow adequate response time to clear
the hopper before the dust reaches the tube sheet and causes blocking of the
inlet to the bags. Not all fabric filters use or need hopper level indicators.
Two types of level indicators are most commonly used (although others
are available). The older of the two, a capacitance probe, is inserted into
the hopper. As dust builds up around the probe, a change in the capacitance
occurs and triggers an alarm. Although these systems are generally reliable,
they can be subject to dust buildup and false alarms in some situations. The
newer system, currently in vogue, is a nuclear or radioactive detector.
These systems use a shielded Cesium radioisotope to generate a radioactive
beam that is received by a detector on the opposite side of the hopper. Ash
intercepting the beam decreases the detector signal and triggers a response.
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-11
-------
This system has two advantages: 1) it does not include a probe that is sub-
ject to dust buildup, and 2) more than one hopper can be monitored by one
radioactive source. Its major drawback is that plant personnel must deal
with a low-level radioactive source, which means adequate safety precautions
must be taken. These detectors are provided with safety interlocks to pre-
vent exposure of plant personnel when maintenance is required.
Hopper level detectors normally should be placed between one-half and
two-thirds of the way up the side of the hopper. As long as hoppers are not
used for storage, this should provide an adequate safety margin. (It should
be remembered that it takes much longer to fill the upper 2 feet of a pyramid
hopper than the lower 2 feet.)
Other indirect methods are available for determining whether the hopper
is emptying properly. On vacuum discharge/conveying systems, experienced
operators can usually tell where the hopper is plugged or if a "rat-hole" is
formed by checking the time and vacuum drawn on each hopper as dust is re-
moved. On systems that use a screw conveying system, the current drawn by
the conveyor motor can serve as an indicator of dust removal. Another simple
method for determining hopper pluggage is through a thermometer located
approximately two-thirds of the way up on the hopper. If dust covers the
probe because of hopper buildup, the temperature will begin to drop, which
signals the need for plant personnel to take corrective action.
3.3 PERFORMANCE TESTS AND PARAMETER MONITORING
The operating characteristics of a fabric filter are such that several
concepts are useful in a performance evaluation. Among these are parameter
monitoring and baseline assessments. These concepts form the basis for good
recordkeeping and a preventive maintenance program aimed at achieving contin-
uous compliance of the controlled source.
Regulatory compliance is determined by a performance test involving the
use of a Reference Method, such as Method 5 or Method 17. In between these
periodic performance tests during operations at maximum and normal operating
conditions, compliance with visible emissions standards also may be deter-
mined by opacity observations performed in accordance with requirements of
Reference Method 9. Because these emission tests represent only a small
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-12
-------
segment of time in the daily operation of the process and fabric filter, the
performance during the emissions test may not be representative of character-
istic daily operations. Nevertheless, these emissions tests do afford the
opportunity to document process and fabric filter operating conditions that
influence performance. By providing a known level of performance, these
values serve as a benchmark or baseline condition for future comparisons with
data collected during routine parameter monitoring and recordkeeping or as
part of diagnostic troubleshooting. The establishment of these baseline
conditions makes it possible for a number of parameters to be compared to
determine their effect on performance. It is the magnitude of these changes
that is important. In addition to the data obtained during emission testing,
baseline conditions may include both pressure and temperature drop across
fabric filter (average as well as before and after cleaning) and cleaning
frequency.
Parameter monitoring, an extension of baselining the fabric filter and
process equipment, forms the basis of diagnostic recordkeeping and preventive
maintenance. Several key parameters are usually monitored to track fabric
filter performance. Generally, parameter monitoring includes both process
and fabric filter data because both are important to fabric filter perfor-
mance. An analysis of these key parameters and a comparison with baseline
values can define many performance problems, indicate the need for mainte-
nance, and define operating trends within the fabric filter (see Section 4).
3.3.1 Performance Tests
The performance test often is the deciding factor for the acceptance of
a new fabric filter, and many agencies require periodic testing (anywhere
from quarterly to once every 3 to 5 years). The initial performance test
certifies that the fabric filter is designed to be capable of meeting the
specifications. These initial performance tests may also include tests with
sections of the fabric filter out of service to meet special requirements of
the permit, specifications, or regulatory requirements. The initial perform-
ance tests may also include inlet tests to establish grain loading, collec-
tion efficiency, and, in some cases, inlet particle size characteristics.
Testing requirements vary from site to site, and they should be estab-
lished in a testing protocol; however, one of two test methods is generally
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-13
-------
specified for determining participate emission rates. These are EPA Refer-
ence Methods 5 and 17 (40 CFR 60, Appendix ). In both methods, a sample is
removed isokinetically (i.e., the linear velocity of the gas entering the
sampling nozzle is equal to that of the undisturbed gas stream at the sam-
pling point) from various sample locations to prevent the -sample results from
being biased. The main difference between the two methods is the location of
the filter in the sample train. The Method 5 sample train uses an external
filter held in a temperature-control hot box. The sample passes through the
heated sample probe and filter into the impinger train for removal of conden-
sable materials (water, acid, and condensible organics). The particulate
emission rate usually will be determined from the probe and filter catch
alone (front-half catch); however, some regulatory limitations specify the
use of both front- and back-half catches (which include the impinger catch
minus water). In most cases, the specified temperature is 248° ± 25°F for
the filter of a Method 5 sample train, but special conditions allow a temper-
ature up to 320° ± 25°F.
Method 17, on the other hand, uses an in-stack filter to capture partic-
ulate. After the filter temperature has been allowed to equilibrate to stack
conditions, the sample is drawn through the nozzle and into the filter. The
sample is then passed through a set of impingers to remove condensibles from
the gas stream.
The two methods often do not provide equivalent results, even when the
flue gas temperature is the same as the hot box temperature. First, Method 5
defines particulate as the material that is captured on a filter at the hot
box temperature (nominally 250°F), even though the temperature of the gas
stream passing through the filter may be substantially different from the hot
box temperature. This is important because many "particulates" are tempera-
ture-dependent, i.e., they exist below a certain temperature, but they may
remain in a gaseous form above a given temperature. Theoretically, particu-
late matter is referenced to a particular temperature. Second, some losses
are normally associated with recovering the particulate from the in-stack
filter of the Method 17 sample train, and an average correction must be ap-
plied (e.g., 0.04 gr/scf for kraft recovery boilers). Lastly, because Method
17 is referenced to the stack temperature, the definition of what constitutes
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-14
-------
"particulate" may be different. Method 17 is usually reserved for particu-
lates or processes that are not temperature-dependent. When this is the
case, the results of both methods are usually in relatively close agreement.
There has been widespread discussion as to which method should be used
when a choice is allowed. Each method can be manipulated to provide the most
favorable emission rate. Method 17 may reflect more accurately the conditions
the fabric filter may encounter; however, Method 5 attempts to standardize
the operating temperature so that differences in temperature and temperature
dependency are minimized.
In addition to overall particulate emission rates, some regulations
limit the emission of fine particles. Such limits require particle size
analysis, either by microscopic methods or by the use of cascade impactors.
Cascade impactors are placed in the stack in a manner similar to the place-
ment of an in-stack filter. An impactor consists of a series of perforated
plates and target or impact stages on which an impact medium (e.g., grease)
is used. As the gas and particulates pass through the impactor, they are
accelerated to higher velocities. The particulate matter has difficulty
staying with the flow streamlines, and its inertia carries it to impact the
target stage. Each stage is sized to capture a predetermined particle size
range at a given flow rate. Calibration curves and corrections to the parti-
cle size ranges are provided by the equipment manufacturers.
There are two problems related to the use of cascade impactors. First,
the gas stream must be sampled isokinetically to avoid skewing of the parti-
cle size distribution. Under-isokinetic sampling (at a probe velocity less
than that of the stack or duct) usually results in a distribution skewed
toward large particles, whereas over-isokinetic sampling undersamples the
large particles. Also, the particle size distribution may have little
bearing on the Method 5 results because of the temperature-dependency of the
particulate. Second, the isokinetic sampling requirement means that the sam-
ple must be drawn at a given flow rate and the flow rate cannot be varied to
maintain calibration of the impactor. Varying the flow rate would vary the
particle size distribution each stage would capture. This usually results in
the use of single-point sampling or an "average" isokinetic rate for the test
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-15
-------
ports, with all the attendant limitations regarding sample representative-
ness. In some cases, multipoint sampling may be carried out, but careful
planning and, for multiple impactors, good flow distribution are necessary.
Opacity is usually monitored during the performance test with an opacity
monitor and/or by Method 9 observations. Several sources (most notably
utilities) have done some work to establish mass/opacity correlations, which
prompt the following general observations. First, mass/opacity correlations
appear to depend on a number of industry- and site-specific factors, including
size distribution, stack path length, and process-related factors. Second,
over a period of time, consistent relationships have been found at a number
of sources (regardless of process load), provided neither the process nor the
fabric filter is experiencing severe malfunctions. The process operation
seems to be the controlling factor in most cases. For example, much of the
data for utility applications suggests that the mass/opacity relationship is
relatively constant for a given source; however, when a condensible partic-
ulate is present, the mass/opacity correlation may not be reliable. Third,
opacity monitor data tend to produce "tighter" or better correlations than
Method 9 when 95 percent confidence intervals are calculated for the correla-
tion. Observer biasing can result from changing background conditions or
from "between-observer" differences; whereas a properly maintained monitor
usually is not subject to biasing problems. Lastly, confidence intervals
tend to become very large at the extreme ends of the curves when mass and/or
opacity is either very low or high. At the low mass/opacity end of the
curve, the relative errors, particularly in test methods, can become substan-
tial. Although the mass loading is very high at the upper end, little change
may occur in opacity. In the opacity ranges of interest to most agencies and
sources, however, the confidence intervals can be quite tight. This opacity
correlation, although not usually used for compliance determinations, can be •
useful in evaluating operation and maintenance, which was the original intent
behind the requirement for continuous emission monitors.
3.3.2 Baseline Assessment
The establishment of baseline conditions for a fabric filter during a
performance test provides a basis for comparison in future evaluations of the
fabric filter. The baseline serves as a reference point, and the types and
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-16
-------
magnitude of shifts from baseline conditions are important in evaluating
fabric filter performance.
Fabric filter baseline conditions generally can be established during
the unit's initial performance test. The following key parameters should be
examined during subsequent performance tests and compared against the base-
line values:
1. Gas volume. If too high, it can blind the bags; if too low, it can
cause dust dropout in the ducts.
2. Temperature. If too high, it can destroy the bags and/or gasket-
ing; if too low, it might cause excursions below the dewpoint.
3. Pressure drop. A pressure drop that is too high indicates poten-
tial bag blinding or high gas flow; one that is too low indicates
bag failure.
4. Dust load. If too high, it may exceed the unit's capacity to
convey the dust from the baghouse; if too low, it may cause ex-
cessive emissions after each cleaning cycle.
5. Particle size. Particles that are too fine can cause blinding of
the bags or excessive emissions.
Baseline conditions should also be established for the process that the
fabric filter controls. Process data may include production weight, raw
material and product feed characteristics, operating temperatures and pres-
sures, combustion air settings, and cycle times (for cyclic processes).
Although the exact effect a change in most of these parameters will have
on performance cannot be predicted, a qualitative evaluation can often be
made when values deviate from baseline conditions, and these deviation values
are useful in parameter monitoring.
3.3.3 Parameter Monitoring
Parameter monitoring usually plays a key role in an overall operation
and maintenance plan, particularly one that stresses preventive maintenance.
Such monitoring also forms the basis for a recordkeeping program that places
emphasis on diagnostics. Typically, daily operating data are reduced to
include only the data on a few key parameters that are monitored. Acceptable
ranges may be established for various parameters (by use of baseline test
data) that require further data analysis, or perhaps some other action if the
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-17
-------
values fall outside a given range. Care must be taken not to rely on just
one parameter or indicator; other factors, both design- and operation-related,
usually must be considered. Typical parameters that can be monitored include
opacity and pressure drop during cake buildup, and gas temperature.
Many sources use opacity levels as the first indicator of performance
changes. In general, opacity is a good indicator and tool for this purpose.
It is not wise to rely on opacity data alone, however, as such reliance can
cause one to overlook problems that can affect long-term performance (e.g.,
hopper pluggage or bridging within bags may not significantly increase opac-
ity, but may eventually decrease the net cloth area and increase pressure
drop).
Another useful parameter is the pressure differential across the fabric
filter. Static pressure drop should be measured periodically to determine
relative changes in dust cake resistance. When reviewing the operating logs,
the operator should look for any increase above the previous operating levels
in the lower (after the cleaning cycle) and upper (before the next cycle
begins) pressure drops across the bags. A gradual increase in resistance can
indicate oil deposits, fine particulate blinding of fabric, or moisture
inleakage. An increase may be tolerated if it is not severe or if a de-
creased ventilation performance does not result from the decreased volume of
gas exhausted. Temperature charts must be monitored to determine the poten-
tial for short-term failure caused by temperature excursions and to detect
inleakage to the fabric filter housing. It is an unfortunate misconception
that short temperature excursions do not cause permanent damage, as the
effects of repeated temperature excursions on tensile strength are cumulative.
A significant decrease in temperature across the fabric filter may indicate
inleakage of outside air, either because of failure of gaskets around open-
ings or the loss of the integrity of the housing.
3.4 RECORDKEEPING PRACTICES AND PROCEDURES
Recordkeeping practices for fabric filters range from none to maintain-
ing extensive logs of operating data and maintenance activities and storing
them on computer disks. The data obtained by parameter monitoring form a
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-18
-------
basis for recordkeeping, as this type of data usually indicates fabric filter
performance. Recordkeeping allows plant personnel to track fabric filter
performance, evaluate trends, identify potential problem areas, and arrive at
appropriate solutions. The magnitude of the recordkeeping activity will
depend on a combination of factors, such as personnel availability, size of
the fabric filter, and the level of maintenance required. For moderately
sized, well-designed, and well-operated fabric filters, maintaining both
daily operating records and maintenance records should not be too cumbersome;
however, records should be limited to key operating parameters only to avoid
accumulating a mountain of unnecessary information.
When setting up a recordkeeping program, one should give attention to
both operating and maintenance records because they are required to provide a
complete operating history of the fabric filter. This operating history is
useful in an evaluation of future performance, maintenance trends, and
operating characteristics that may increase the life of the unit and minimize
emissions. Even though recordkeeping programs are site-specific, they should
be set up to provide diagnostic and troubleshooting information, rather than
merely for the sake of recordkeeping. This approach makes the effort both
worthwhile and cost-effective.
Other supplementary records that should be maintained as part of the
permanent file for operation and maintenance include all baseline assessments
that include both process and fabric filter operating data and data from
emission tests. A spare parts inventory listing also should be maintained,
with periodic updates so that parts may be obtained and installed in a timely
manner.
3.4.1 Operating Records
As mentioned previously, the specifics on what parameters will be moni-
tored and recorded and at what frequency will be largely site-specific.
Nonetheless, the factors that are generally important in parameter monitoring
will also be the ones recorded as part of a recordkeeping program. Such data
would typically include the process operating rate, differential pressures,
temperatures, and opacity monitor readings. These data probably should be
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-19
-------
gathered at least once per shift. The greater the frequency of data gather-
ing, the more sensitive the operators will be to process or fabric filter
operational problems, but the amount of data to manipulate also increases.
The optimal frequency may be every 4 hours (twice per shift). In the event
of sudden and dramatic changes in performance, this allows relatively little
time to evaluate the data to determine the cause of the change. Shorter
intervals might be required on highly variable sources.
In addition to the numerical values of the operating parameters, a check
list should be included to confirm operation of the cleaning system, hopper
systems (or other dust-removal systems), the absence of audible inleakage,
and the other general physical considerations that can adversely influence
fabric filter performance.
3.4.2 Maintenance Records
Maintenance records provide an operating history of a fabric filter.
They can indicate what has failed, where, and how often; what kind of prob-
lems are typical; and what has been done about them. These records can be
used in conjunction with a spare parts inventory to maintain and update a
current list of available parts and the costs of these parts.
The work order system provides one of the better ways to keep mainte-
nance records. When properly designed and used, this system can provide
information on the suspected problem, the problem actually found, the correc-
tive action taken, time and parts required, and any additional pertinent
information. The system may involve the use of triplicate carbon forms or it
may be computerized. As long as a centralized system is provided for each
maintenance activity, the work order approach usually works well.
Another approach is to use a log book in which a summary of maintenance
activities is recorded. Although not as flexible as a work order system
(e.g., copies of individual work orders can be sent to various appropriate
departments), it does provide a centralized record. This type of maintenance
record is probably better suited for the facility that uses smaller fabric
filters.
In addition to these centralized records, a record should be maintained
of all periodic checks or inspections. These should include the weekly,
monthly, semiannual, and annual checks of the fabric filters that make up
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-20
-------
part of a preventive maintenance program. Specific maintenance items identi-
fied by these periodic inspections should be included in the recordkeeping
process. The items to be checked are discussed in more detail in Sections 4,
5, and 6.
3.4.3 Retrieval of Records
A computerized storage and retrieval system is ideal for recordkeeping.
A computer can manipulate and retrieve data in a variety of forms (depending
upon the software) and also may be useful in identifying trends. A computer-
ized system is not for everyone, however. The larger the data set to be
handled, the more likely it is that a computer can help to analyze and sort
data. For a small source with a fabric filter that presents few problems and
that has a very manageable set of operating parameters to be monitored, a
computer system could be very wasteful (unless computing capability is al-
ready available). Also, it is sometimes easier to pull the pages from a file
manually, do a little arithmetic, and come up with the answer than to find
the appropriate disks and files, load the software, and execute the program
to display "the answer."
Retention time is also a site-specific variable. If records are main-
tained only to meet a regulatory requirement and are not used or evaluated,
they can probably be disposed of at the end of the statutory limitation
(typically 2 years). Some would suggest that these records should not be
destroyed because in the event of premature failure of a fabric filter (or
process), the data preserved in these records could be used as an example of
what not to do. On some fabric filters in service today, records going back
10 to 12 years have been kept to track the performance, cost, and system
response to various situations and the most effective ways to accomplish
things. These records serve as a learning tool to optimize performance and
minimize emissions, which is the underlying purpose of recordkeeping. Some
of these records may very well be kept throughout the life of the equipment.
After several years, however, summaries of operation and maintenance activi-
ties are more desirable than the actual records themselves. These can be
created concurrently with the daily operating and maintenance records for
future use. If needed, actual data can then be retrieved for further evalua-
tion.
SECTION 3-FABRIC FILTER PERFORMANCE MONITORING
3-21
-------
SECTION 4
PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS,
AND PROBLEM SOLUTIONS
Many facilities that have installed fabric filters have done so with
expectations of high collection efficiency, reasonable energy requirements,
good long-term performance with reasonable bag life, and low maintenance
requirements. Unfortunately, failure of the equipment to meet these expec-
tations (because of poor design and installation, inadequate maintenance, or
operation that is inconsistent with design constraints) often results in non-
compliance and/or maintenance problems. Personnel responsible for evaluating
the performance of a fabric filter should be familiar with the equipment
design, the principles of operation, the importance of certain process param-
eters and their effect on performance, and maintenance requirements. This
section presents discussions on the use of key parameters in monitoring per-
formance, diagnosing or troubleshooting problems, and providing corrective
actions for the most common problems. Because fabric filters are applied to
a wide variety of sources, site-specific factors play an important role in
the overall performance of these control systems.
4.1 PERFORMANCE EVALUATION
As discussed in Section 3, several key operating parameters should be
monitored and recorded to alert operators to current performance conditions.
To be of greatest value, however, these parameters must be compared against
initial design conditions or baseline values to determine their acceptabil-
ity. For example, measurement of temperature and pressure drop, two of the
more common parameters, has little meaning if the design and baseline values
of these parameters are not known. The temperature limits may be set by
dewpoint conditions and by bag type, both of which are part of the design and
baseline criteria data set. Without these data, the value recorded for
temperature has relatively little meaning. A similar case can be made for
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS
4-1
-------
pressure drop. Thus, a comparison of the key operating parameters with
design or baseline values provides an awareness of what values are "normal"
or fall within the range of acceptable performance.
Although a single data set is useful for evaluating operating conditions
and performance, it generally cannot be used to evaluate trends in the fabric
filter performance. Evaluating long-term performance, minimizing maintenance
costs, and providing optimum bag life require the maintenance and evaluation
of daily records so that any sudden or gradual changes in the parameters can
be determined. Key process and control equipment operating parameters are,
of course, of greatest interest in such an evaluation. With fabric filter
systems, the parameters that should be reviewed to determine trends are
opacity, pressure drop, temperature, pressure drop and temperature cycles (if
continuous strip chart recorders are used), and such key process parameters
as production rate. Fan speeds and motor currents also could be included in
this evaluation. When evaluated in light of the gas temperature, the fan
speed and motor current can be indicative of the gas volume moving through
the system. Maintenance records are also very useful in evaluating long-term
performance, in establishing performance trends, and in identifying failure
patterns. For example, records of bag failures, their location, and the type
of failure could reveal a design or operating problem when used in conjunc-
tion with daily operating records. How the records are arranged and what
data are included are site-specific decisions.
4.2 DATA COLLECTION AND COMPILATION
For most fabric filter applications, the two most useful operating
parameters are the opacity and the pressure drop across the filter material.
In some applications, temperature data will be important for evaluation of
the impact of high temperature excursions or condensation. The frequency for
collection of these data will depend on several factors, but as a general
rule, these data should be checked once a day. Continuous strip-chart
recorders for opacity, pressure drop, and temperature can be very useful for
indicating daily trends. The use of continuous strip-chart recorders is
often limited to larger sources, however. Opacity monitors also tend to be
used only in certain applications and at large industrial sources. Most
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-2
-------
small sources that use fabric filters have no opacity monitors. All sources,
however, have means of monitoring temperature, and when fabric filters are
applied to high-temperature sources or in situations where temperature prob-
lems may occur, the inlet gas temperature should be monitored continuously.
In the absence of continuous monitors/recorders, visible emission char-
acteristics and onsite instrumentation must be observed periodically and the
results evaluated. Opacity observations are very useful at most applications
because opacity plumes at a properly operated and maintained fabric filter
are generally very low, except when a condensible plume is present. A
relatively continuous elevated opacity level can be indicative of the pres-
ence of major leaks and tears in the filter bags. Pinhole leaks are also
usually discernible by an increase in opacity after cleaning of the bag(s).
These kinds of plume characteristics are generally discovered by continuous
observation of the plume as opposed to once every 15 seconds as required by
EPA Reference Method 9. Whereas the use of Method 9 for determining average
opacity is sufficient for enforcement purposes, changes in opacity that
result from minor leaks may be missed when this method is used. In general,
continuous observation of the plume to note any changes is better suited for
evaluating maintenance requirements, as problems in certain rows or modules
can be identified by this method.
The pressure drop across the fabric filter gives an indication of the
resistance to gas flow and cleaning effectiveness. The pressure drop usually
varies with the square of the gas volume flow through the fabric, but it will
also vary with the thickness of the dust cake and the amount of material
permanently retained by the fabric filter. This value will depend on various
factors. The pressure drop of a fabric filter, however, generally falls
within a "typical" range, and it is this range that is important. The
recorded value should fall within the general operating range for the unit.
Any changes in the pressure drop, whether gradual or sudden, may indicate the
need for maintenance. If the cleaning cycle is initiated by a specified
pressure drop, however, the pressure drop will not change, but the time
between cleaning cycles will be shortened. When a large number of fabric
filters must be evaluated, forms may be printed that include the typical or
baseline values so that an immediate comparison can be made. For large,
SECTION 4-PERFORMANCE EVALUATION. PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS
4-3
-------
multicompartmented filter systems, recording the pressure drop across indi-
vidual modules may not be necessary because pressure drop tends to equalize
across all the modules.
For those units on which operating temperature is of particular concern,
the use of continuous strip-chart monitors is highly recommended. Sometimes
bag damage is not evident until days or weeks after a temperature-related
incident. This can be troublesome to maintenance personnel because failure
to detect the cause of deteriorating bags can result in unusually high
maintenance costs. Although both inlet and outlet monitors are recommended,
measurement of only the inlet gas temperature is usually sufficient. Tem-
perature readings recorded during the acquisition of other data (opacity,
pressure drop, production rate, etc.) are usually of little use by them-
selves, since they are not continuous.
Maintenance records are also useful in evaluating fabric filter perform-
ance. A record of bag failures and/or bag replacement can be especially
helpful. In a typical application with newly installed bags, random bag
failures shortly after startup is not uncommon. These are usually caused by
an occasional defective bag and by installation problems. After these
failures occur during the shakedown period, bag replacement requirements are
expected to be minimal until the bags near the end of their useful lives.
Records of bag replacement location, however, may reveal the presence of
failure patterns resulting from design or operating practices. The existence
of such a pattern may suggest a possible cause and solution that will improve
performance and reduce maintenance costs in the long run. A typical bag
replacement record is shown in Figure 4-1.
Another characteristic that bears examination is the physical property
of the dust and any associated changes that may have occurred. Although
site-specific factors control the characteristics of the dust to be con-
trolled, two general characteristics that can influence fabric filter per-
formance are particle size distribution and the adhesive characteristics of
dusts. Changes in particle size distribution may increase abrasive wear if
the particles increase in size. On the other hand, a shift to a smaller
particle size range may increase penetration (bleed-through) and blinding.
Changes in process operating characteristics can sometimes cause significant
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS
4-4
-------
BH»
MOD.*
5"
DATE
18
17
16
14
15
12
11
10
9
8
7
5
4
3
2
1
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
•oo
ooo
ooo
ooo
OOOO'
•ooo
OOOO"
OOOO"
OOOO"
OOOO"
*
OOOO
OOOO'
OOOO6
•OOO'
OOOO3
OOOO2
•OOO'
ooo
ooo
ooo
ooo
ooo
ooo
oo
oo
oo
o
o
ooo
••
ooo
ooo
18
17
16
13
14
13
12
11
10
C t>
Key:
B - Broken/Hole
No Letter - Ring
C - Collapse
Figure 4-1. Typical bag replacement record.
(Courtesy of Mikro-Pul Corp.)
4-5
-------
shifts in particle size. Changes in adhesive characteristics can also result
from variations in process operation conditions (e.g., some combustion
sources can produce "sticky" carbon particles if combustion characteristics
are poor) or fluctuations in temperature that produce dewpoint problems.
Where such changes are possible, routine monitoring of dust characteristics
may be prudent to prevent excessive or unexpected maintenance problems.
4.3 PROBLEM DIAGNOSIS
The two major categories of operation and maintenance problems are
1) problems that can affect fabric filters regardless of type, and 2) prob-
lems that are characteristic to a particular cleaning system design. The
first category includes fabric failure, dust discharge problems, corrosion,
poor or improper maintenance considerations, and other problems that are
common to nearly all fabric filter types. The discussion of problems in the
second category is presented by system design—shaker, reverse-air, or
pulse-jet. Some hybrid systems will not conform to these criteria, and
site-specific design factors must be considered. Most failure modes, how-
ever, are addressed here.
4.3.1 Fabric Failures
The factors that contribute to fabric failures include improper instal-
lation, high temperatures, condensation, chemical degradation, high air-to-
cloth (A/C) ratio, high pressure drop, and bag abrasion. Each of these is
discussed separately.
Installation--
The first step in achieving the expected performance from the fabric
filter is proper installation of the bags in accordance with the guidelines
provided by the bag manufacturer and the equipment vendor. Because these
guidelines are not always available to maintenance personnel, training of
maintenance personnel in proper installation procedures is very important.
Reasons for lack of training vary, but generally they result from lack of
vendor-supplied training and maintenance manuals and turnover in maintenance
personnel. The latter creates a situation that necessitates almost continual
training. Common problems resulting from improper installation of the bags
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-6
-------
include leaks around seals, improper bag tensioning, and damage to the bags
during handling.
Failure of a few new bags is to be expected in a new or rebagged fabric
filter, even during normal operating conditions, as a result of occasional
manufacturing defects and improper handling during installation. After these
few initial failures, however, if the system is properly designed and oper-
ated, the occurrence of such failures should be at a very low level until
near the end of the bag life, when failures are likely to start increasing.
If installed improperly, however, bags may continue to fail long after the
initial installation and long before they approach the end of what should be
their normal life.
To some extent, the design of the fabric filter can influence the extent
of bag damage during installation. Systems that provide good access and are
designed with maintenance considerations in mind reduce the likelihood of bag
damage. The poor access design of a reverse-air fabric filter (see Figure
4-2) is not conducive to proper installation and maintenance. Examples of
designs that facilitate bag replacement or installation are in pulse-jet
systems with top-loading bags or reverse-air or shaker systems with a maximum
"bag reach" of two or three bags. These designs allow maintenance personnel
to disturb only a minimal number of "good" bags. Obviously, little is gained
if the replacement of one or two bags results in the damage and life-short-
ening of several others.
In reverse-air and shaker-type fabric filter installations, damage often
occurs when the bags are being hung. Access to the bag support and tension-
ing mechanisms may be difficult and cumbersome, and personnel may prefer to
hang all the bags at once and then attach them to the tube sheet after all of
them are suspended. The tendency, however, is to tie the bags out of the way
as they are hung in the enclosure. (See Figure 4-3). Some fabric types can
withstand this treatment with few problems, whereas others (most notably
fiberglass) cannot. If the fabrics have poor abrasion resistance, tying the
bags can cause leaks to occur wherever a crease is formed. Efforts should be
made to tension these bags properly as they are installed, as this adjustment
is sometimes difficult after all the bags have been hung. Care also should
be taken to avoid stepping on the bags as they are taken out of their cartons
or when they are laid on the floor prior to installation.
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-7
-------
of exc
4-8
-------
Figure 4-3. Tied-off fiberglass bags during bag replacement.
(This may result in damage to new bags during installation.)
(Courtesy of PEI Associates, Inc.)
4-9
-------
Fabric filters with pulse jet cleaning systems (or any other design
where filtration occurs on the outside and a cage is used for bag support)
generally have shorter bags, which are a little more manageable. Most of the
damage during installation of these bags occurs when they are placed on the
cages. These bags generally fit snugly, and damage may result from im-
properly sized bags or sharp edges on the cages. When a number of bags are
being installed, the bags are generally placed on the cages and stacked prior
to their installation. This can result in bag damage unless special care is
taken. These bags can also be damaged if they must be slid through a tube
sheet and the fit is too tight. The "nip" or amount of the bag that can be
drawn away from the cage for a correct fit is about 1/4 inch.
In summary, failure to take appropriate precautions to safeguard the
bags during the installation process may result in excessive maintenance due
to bag failure or reduced bag life.
High Temperature--
High temperatures are not a consideration with many fabric filter
applications; however, in those that operate above 150°F, the effects of
temperature on the fabric must be considered. High temperature breaks the
polymer chains in most commercially available fabrics, which causes loss of
strength and reduces bag life. The effects are different on high-temperature
fiberglass. The high temperatures attack the finish that has been applied to
the fiberglass to reduce fiber-to-fiber abrasion, and when this finish is
destroyed, the bag can abrade itself and self-destruct. Sometimes it takes
several days or weeks before these bags begin to fail. The fabric type is
chosen on the basis of expected temperature ranges, and care must be taken to
provide an adequate margin for error. Temperature monitors and alarms are
often used to avoid high temperature excursions. Excursions above the
recommended temperature limit generally shorten bag life considerably;
however; the closer the actual operating temperature is to the fabric's
temperature limit, the shorter the bag life will be.
Condensation—
Condensation of moisture and/or acid gases is generally associated with
reduced temperatures within the fabric filter. Condensation of moisture or
acid mist on the bags tends to alter the adhesion characteristics of the dust
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS 4-10
-------
cake on and within the fabric structure, and "mudding" or blinding of the
bags may occur because the cleaning system cannot remove this dust. This
usually increases the pressure drop, and more fan energy is required to
overcome the added resistance through the dust cake. Such conditions may
occur near the walls of the unit when warm, moist, or acidic gases pass
through a cool or cold fabric filter that has not been preheated. (For
example, it is often advisable to preheat the fabric filter unit at an
asphalt plant prior to introducing wet aggregate to the dryer.) Condensation
may also occur if moist gases are not purged from the unit before it is shut
down. When the temperature is allowed to cool below the dewpoint at the end
of a production run, moist or acidic gases should be purged to prevent
condensation on the walls and the bags within the fabric filter.
Chemical Degradation—
Chemical resistance refers to the fabric's ability to withstand acidic
or alkaline conditions. Fabrics are rated according to their chemical
resistance, but care must be exercised with regard to these classifications
because certain fabrics are more susceptible to some chemical species than to
others. Fiberglass bags generally are rated as having good acid resistance;
however, the use of fiberglass bags in atmospheres with appreciable quan-
tities of hydrogen fluoride would not be advisable. Nomex^is generally
rated as having fair acid resistance. It is also generally known for its
good moisture and S00 resistance; when both water and SCL are present,
however, sulfurous and sulfuric acid mist is formed, the aramid structure of
Nomex is attacked, and the fabric loses its strength. Individually, water
vapor or S02 does not present a problem to Nomex, but in combination they can
result in costly bag failure.
High A/C Ratio-
High A/C ratio generally results from an increase in gas volume moving
through the system or the installation of an undersized system. The cleaning
system type generally controls the range of acceptable A/C ratios; the
lower-energy cleaning systems (reverse-air and shaker) use lower A/C ratios.
Other factors, such as dust loading and particle size distribution, also
influence the design A/C ratio. In general, the higher the A/C ratio, the
higher the operating pressure drop. Excessively high A/C ratios, however,
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS 4-11
-------
can result in very high pressure drops, and bag abrasion is increased because
the particles impact the bags at higher face velocities. Bag blinding or an
increase in the residual dust loading after cleaning also may occur because
both the increased pressure drop across the bag and the increased velocity
allow dust to penetrate into the fabric, where the cleaning system is unable
to remove it. This will cause a gradual increase in pressure drop across the
bags. The net result is generally an increase in energy requirements to
maintain gas flow and a decrease in bag life.
High Pressure Drop—
As noted previously, high pressure drop can be a symptom of high A/C
ratios. It can also occur when the cleaning system fails or when little or
no cleaning energy is supplied to remove the dust cake from the bags. The
greater thickness of the dust cake increases the resistance to gas flow,
which in turn is reflected as an increase in pressure drop across the bags.
High pressure drop also can result from bag blinding or condensation in the
bags. Although it is usually a symptom of some other problem, high pressure
drop itself may cause other problems. First, the greater resistance to flow
tends to decrease gas flow through the fabric filter and can lead to fugitive
emissions from the emission source. An increase in expended energy is
required to maintain gas flow. Second, the greater differential pressure
between the dirty and clean side of the fabric provides a larger amount of
energy to draw particulate matter into the weave of the bag, which can lead
to more abrasion damage within the bag and shortened bag life. Lastly, a
very high pressure drop (10 to 14 in.) may cause the bag to be unable to
withstand the pressure differential and to tear at points where the bag's
strength has been reduced. In most fabric filters, only a few affected bags
can lower the pressure drop and allow significant quantities of gas to pass
through the fabric filter untreated.
Although high pressure drop is usually a symptom of other problems and
should be treated as such, it should not be ignored. Even if the related
problem does not shut the fabric filter down, the high pressure drop will
lead to higher energy costs, reduced bag life, and increased maintenance
costs.
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-12
-------
Bag Abrasion-
Bag abrasion may be caused by contact between a bag and another surface
(e.g., another bag or the walls of the fabric filter) or by the impact of
higher-than-average gas volumes and particulate matter loading on the bags.
Bag-to-bag contact can be a problem in nearly every type of fabric filter if
the bags are not installed properly. Such contact may eventually wear a hole
in the bag, and the resulting jet of gas flow through the hole will gradually
enlarge it. On bags that collect dust on the inside, a hole may cause a
high-velocity jet to impinge upon an adjacent bag and also eventually wear a
hole in it.
Blast plates or diffusers (and sometimes precleaning devices) are
recommended for many fabric filters. The purpose of these devices is to
reduce the quantity of large particles that strike the bags and, along with
long thimbles on shaker and reverse-air systems, to help minimize the wear on
the bottom of the bag. Because of their size and weight, these large par-
ticles have great inertia, which allows the particles to strike the bag at an
angle and eventually damage the bags. These particles have a tendency to
stratify in the inlet of the fabric filter because their inertia reduces
their ability to follow gas streamlines. It is not unusual to find abrasion
problems in the bags on the side opposite the inlet. On most bags the
greatest abrasion occurs within 18 to 24 inches from the bottom of the bag.
Diffusers, such as the one shown in Figure 4-4, tend to help reduce the
problem, and the diffuser should be checked periodically for wear.
Holes in the bags usually cause an increase in opacity, if small par-
ticles are present, and they also may cause a reduction in pressure drop.
Pinholes are usually covered easily by the dust cake; thus, opacity increases
after the bag is cleaned. This increase in opacity is relatively short,
however, and diminishes as the pinhole is covered again. Tears or holes in
the bags may or may not be covered by the dust cake, depending on their size
and the pressure drop across the bags. The opacity generally will not
decrease quickly or substantially, however, because the hole(s) may allow a
significant quantity of material to pass through the system. Thus, opacity
can be an indicator of the relative magnitude of any holes formed by abra-
sion.
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-13
-------
Figure 4-4.
Example of a diffuser for deflecting large particles
from the gas stream.
(Courtesy of PEI Associates, Inc.)
4-14
-------
4.3.2 Dust Discharge Failures
Hopper pluggage can cause serious problems in a fabric filter. Regard-
less of the reason (cooling of the dust, inleakage, failure of the discharge
system operation, or simply using the hoppers for storage), failure to remove
the dust from the hopper usually results in having to open up the hoppers to
clean them out. This can result in the type of fugitive emissions illustrat-
ed in Figure 4-5. The fugitive emissions generated by a single cleaning out
of the hoppers nay be greater than the emissions emanating from the fabric
filter outlet for an entire year. Therefore, minimizing the occurrences of
hopper pluqgage by emptying hoppers continuously or frequently is very
important.
Many dusts flow less easily when they are cold than when they are warm.
Thus, insulation, hopper heaters, and continuous dust removal may be neces-
sary to minimize the hopper pluggage problems. The effects of hopper plug-
gage are not always immediately obvious. As the dust builds up, dust re-
suspension may increase, as most fabric filter inlets enter through the
hopper. This increase in resuspended material will increase the particulate
loading on the bags, and it also may cause an increase in the pressure drop
across the bags. When the dust buildup in the hoppers reaches a certain
height, some bags may be partially or completely blocked from the gas flow,
which increases the gas flow (A/C ratio) for the remaining bags and further
increases the pressure drop. Eventually, all gas flow may be blocked from
the hopper inlet. Dust buildup in and around the bags can be a problem,
particularly as condensation occurs when the dust is cooled. This can lead
to a condition similar to bag blinding.
4.3.3 Shaker Cleaning System Failures
Because gas flow from the fabric filter must be cut off before the
shaker cleaning system can be operated, shaker-type fabric filters are either
modularized or they are applied to intermittently operating sources where gas
flow can be stopped so the shaking action can be effective. Several problems
are characteristic of shaker-type fabric filters.
Shaker motors can be installed inside or outside of the fabric filter
housing depending on the temperature and corrosive conditions in the gas
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS
4-15
-------
Figure 4-5. Fugitive dust emissions resulting from the opening of hopper
access doors to clean out clogged hoppers.
(Courtesy of PEI Associates, Inc.)
cr>
-------
stream. These small motors (usually less than 5 horsepower) are usually
installed outside the housing and are wired into a control circuit that may
be manually or automatically activated. Operation of these motors is easily
verified. The operation of internally mounted shaker motors, however, can be
difficult to evaluate. Failure of the shaker motor may (and in many cases
does) lead to excessive dust cake buildup on the bags and an increase in
pressure drop. In some applications, when the gas flow is stopped by closing
the dampers, the dust will slide off the bag. In most applications, however,
the shaker system is needed for adequate removal of the dust and maintenance
of a reasonable pressure drop.
The shaker linkages must be maintained in a manner that allows the
energy provided by the shaker motor to be distributed through the shaking
system to the bags. Because these systems are mechanical, periodic lubri-
cation, checking for wear or loose parts, and replacement of broken parts are
required to maintain their cleaning effectiveness. The only way to evaluate
this system is to watch it in operation to ascertain that all the bags are
being cleaned at approximately the same intensity. Watching the system
operate may reveal that certain modules or certain shaker bars are not being
moved through the correct amplitude. These sections have higher resistance
to flow, and the gas is forced to flow through the bags having less resis-
tance to equalize the pressure drop. Although the overall pressure drop may
increase somewhat, abrasion and blinding damage may occur in the bags being
cleaned more effectively by the shaker system.
The third problem in fabric filters with shaker cleaning systems con-
cerns bag tension. Bag tension changes with the age of the bag and with the
amount of material collected on the dust layer, and it is usually expressed
in the number of pounds of force applied to the top of the bags. Thus, bag
tensions are usually adjusted by some arrangement at the top of the bag.
Bags that are too tight may not transfer the shaker energy effectively and
may be damaged during shaking. Bags that are too loose may sag on the
tubesheet, and bag abrasion may result from the bag being placed in the gas
stream or being contacted by the thimble or other bags. Loose bags also may
not use the cleaning energy effectively and may block the flow of dust out of
the bags if they sag, fold, or close off above the tubesheet. Proper tension
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS
4-17
-------
allows the dust to flow out of the bag without sagging problems or problems
in the transfer of shaker energy.
Problems can occur in the bag hanging mechanism if bag tensioning is not
proper. Some systems use chains or threaded bolts that attach the shaker bar
to the top (metal cap) of the bag. In other cases, the bags have a tongue
that is threaded through a clip, and friction is used to keep the bags on
their hangers. Maintenance personnel must properly install the bags to
ensure that they remain attached at the top and that they won't fall down or
lie on the tubesheet. When fabric filters are used to control dense dusts
(e.g., in the metals industries), the bags sometimes fall because they were
installed improperly or because bag tension and/or cleaning efficiency were
inadequate to remove the dust from the bags. When bags become heavily
ladened with dust, they will pull away from the attachment mechanism or cause
this mechanism to break. Because the bags that lie on the fabric filter
floor are essentially out of service, the actual A/C ratio, the pressure
drop, bag wear, and maintenance costs all increase.
4.3.4 Reverse-Air Cleaning Systems
Like the shaker cleaning system, the reverse-air system is a low-energy
system that cannot function properly if gas flow is present in the module or
area being cleaned. The damper systems for fabric filters with this cleaning
mechanism tend to be more complex than those for the shaker system because a
reverse flow of gas is used to collapse the bag, to break and release the
dust cake, and to allow it to be collected and removed from the fabric
filter. Failures in this type of filter system are most often related to the
improper functioning of the cleaning system.
Several types of reverse-air cleaning system designs are available.
Some use a separate reverse air fan, and others do not. Despite the partic-
ular design, the gas flow must be stopped in the module so that cleaning may
take place. This requires a positive seal on the reverse-air isolating
damper (a poppet damper is often used). Without proper sealing, the bags may
not collapse properly and the cleaning action may be ineffective. Unlike the
other cleaning systems, relatively little energy is available to clean the
fabric, as the reverse flow of gas through the bags is usually small compared
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS
4-18
-------
with normal, on-line gas flow. Over a period of time, the overall pressure
drop will gradually increase because of buildup on the bags.
Failure of the isolation dampers is usually easily detected, as the
actuators are generally pneumatically or hydraulically operated and the
movement of the piston is visible. Too little movement of the piston usually
indicates that the damper is not sealing properly. Symptoms of problems are
similar to those for the reverse-air supply dampers. In some situations, the
failure of the damper system can be detected by a missing spike and subse-
quent decrease in pressure drop after the affected module comes off line for
cleaning. Moisture and oil in the compressed-air supply lines can cause
blockage during freezing weather and result in the failure of these pneu-
matically operated systems. Damper operation failures, however, usually
result from failures of the controlling timers or pressure drop sensors that
are used to activate the cleaning cycle at certain intervals or at certain
pressure-drop thresholds.
Buildup of materials around the dampers or deformation of the dampers or
their seals can cause problems with proper isolation of a compartment for
cleaning. Symptoms of this problem ere similar to those for a malfunctioning
damper system, and they may register on the continuous pressure drop recorder.
The major difference is that the damper would appear to be functioning.
Confirmation of poor damper sealing is only possible by internal examination
of the equipment, and even internal inspection of the damper system may be
inconclusive because the system must be cooled sufficiently for safe entry.
An internal inspection, however, may indicate the presence of light leaks,
warped dampers and seals, or buildup or wear of the dampers caused by mate-
rial passing through the fabric filter. The damper operation and seal should
be checked periodically as part of a preventive maintenance program.
Proper bag tension is essential to bag cleaning. Just as it was impor-
tant for the bags to be properly tensioned for the shaking action to be
effectively transmitted to the bag, attention to bag tension is necessary to
obtain the proper collapse and flexing of the dust cake for its removal from
the bags. Bags that are too tight may not collapse enough to allow effective
flexing of the dust cake. Too much tension can also damage the fabric. On
the other hand, insufficient tension of the bags may allow the bags to
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS 4-19
-------
collapse to the point where the bag is closed down during the reverse-air
cleaning cycle (even when anticollapse rings are used). Loose bags also may
suffer abrasion due to the bag being sucked down into the thimble. It is
recommended that thimbles be rounded and free of sharp edges to prevent tears
if this should occur.
Proper bag tension is a function of attention to detail during the
initial installation. Bags must be hung properly, without damage, if the
proper life expectancy is to be achieved. Bag tension will vary with the age
of the bag and also within any given cleaning cycle as material builds up on
the bags. Poor bag tension can increase bag wear, cause high pressure drop,
and shorten bag life.
Corrosion also can be a problem in this type of fabric filter. In some
applications, most notably where acid dewpoint conditions have not been ade-
quately considered, corrosion of the metal anticollapse rings has resulted in
abrasion and wear of the bag at the site of bag ring contact as shown in
Figure 4-6. Sometimes fuels or process parameters can be modified to reduce
potential corrosion. Special alloy metals or coatings also can be used to
minimize or eliminate corrosion problems.
4.3.5 Pulse-Jet Cleaning Systems
Pulse jet fabric filters are widely used because of their smaller size
and because their higher available cleaning energy allows for higher A/C
ratios. Despite the attractiveness of their lower initial costs, however,
these bags have their limitations and potential problems because of the
higher energy required to operate these systems.
The higher A/C ratios on this fabric filter type increase the potential
for fabric abrasion. Therefore, greater efforts should be made to minimize
other, often-overlooked, abrasion-related failures.
Typically, the bags in a pulse-jet fabric filter are suspended from a
tubesheet and supported by a cage. This single-point method of attachment
allows the bag to move around during normal operation. One source of bag
abrasion is bag-to-bag contact due to improper installation, poor alignment
of the bag/cage assemblies with the tubesheet, or bent/warped cages. The
rubbing together of the bags (usually at the bottom) can wear a hole in one
or more of the bags.
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS
4-20
-------
Figure 4-6. Example of bag wear caused
by corrosion of metal anticollapse rings.
(Courtesy of PEI Associates, Inc.)
4-21
-------
The misalignment of bag/cage assemblies can also cause other problems.
In some designs, the misalignment of the cage will prevent proper sealing of
the bag with the tubesheet. This may allow some of the dust to bypass the
filter area, which decreases performance but probably causes little or no
change in pressure drop. Particularly abrasive dust has been known to wear
the bags and the tubesheet so severely at the point of the leak that achiev-
ing an adequate seal may be impossible without replacing the tubesheet.
Another abrasion-related problem concerns the condition of any baffle or
blast plate that may be installed at the inlet of the fabric filter. The
purpose of this device is to "knock down" the heavier particles and to
distribute flow such that the larger particles do not strike the bottom of
the bags opposite the inlet. Not all designs are equipped with a blast
plate, which should bring the gas flow below the bottom of the bags. When
failure of the bags occurs within about 18 inches of the bottom on the side
opposite of the inlet, the presence and integrity of the blast plate or
diffuser plate should be checked. Although other problems with the cleaning
system can lead to increased bag wear and poor performance through higher
pressure drop, these problems tend to be more indirect in nature.
The design and operation of the pulse-jet system generally call for
on-line cleaning, which requires the availability of considerably more clean-
ing energy to remove the dust from the bags (in addition to the higher A/C
ratios normally encountered with this design). Failure to provide this
energy will generally show up quite readily as an increase in pressure drop
because a relatively small cloth area is handling a large gas flow. Several
components can contribute to such a problem.
The compressed-air supply must be able to provide a pressure of between
90 and 120 psig to clean the bags effectively. Compressed-air requirements
for short bags (6 to 8 ft) may be lower (say 60 to 90 psig); whereas for 14
ft bags, pressures of 120 to 140 psig may be necessary for adequate cleaning.
The pressure must be high enough to clean the entire length of the bag during
the pulse, but not so high that it damages the upper portion of the bag.
Insufficient cleaning of the bag may gradually increase pressure drop and
reduce the useful bag life. Too low compressed-air pressure, which is
usually more common than excessive pressure, may be caused by wear of the
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-22
-------
compressor rings, leakage of diaphragms, or excessive draining of the reserve
of the compressors by other equipment tied to a common supply line.
The leakage around a diaphragm, which can usually be detected audibly by
the absence of the resounding "thud" that typically characterizes proper
operation of the pulse-jet system, affects the cleaning effectiveness for all
the bags. Although it may take several hours or several days, the pressure
drop usually will increase eventually if the leak is severe enough.
Failure of the solenoid(s) or the timer circuit may cause one or more
rows not to be cleaned. Effects on fabric filter performance may range from
indiscernible to complete cutoff of gas flow, depending upon the percentage
area of the bags affected and the dust characteristics. Both mechanical and
electronic timers are still in use, and both have certain advantages and
disadvantages. An electronic timer is shown in Figure 4-7; solenoids that
are activated by the timer are shown in Figure 4-8. Both types must be kept
in a dust-free, dry environment and relatively free from the shocks and jolts
that can accompany normal operations. Solenoid failures affect the row that
has experienced the failure whereas timer failures tend to affect most, if
not all, of the fabric filter system.
When the timer activates the solenoid that opens the diaphragm at the
end of the pulse pipe, the force of the compressed air entering the pulse
pipe and discharging into the bags places considerable stress on the pulse
pipe. In some instances, the force is sufficient to break the attachment at
the other end of the pipe (usually a bolt and nut), which allows the pipe to
bounce around inside the fabric filter when the row is cleaned. Several
problems may result. First, the pulse pipe may not be properly aligned to
provide effective cleaning to that row. Second, the alignment may be such
that the pipe openings are aimed directly at the bags and can blow holes in
them. Lastly, the loose pipe may damage the tubesheet or even the fabric
filter enclosure, which would necessitate additional repairs. The sound of a
loose pulse pipe is usually unmistakable, as it moves around whenever the
pulse-jet compressed-air is fired into that pipe.
Although all of these problems are relatively common in most pulse-jet
systems and may produce bag abrasion or shorten bag life, the one problem
that seems to recur with greatest frequency is the presence of water and/or
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-23
-------
Figure 4-7.
(LED's at each
Electronic timer circuit board for a pulse-jet filter.
position indicate when the circuit has fired a signal'to
open the pulse-pipe diaphragm.)
(Courtesy of PEI Associates, Inc.)
4-24
-------
Figure 4-8. Pulse-jet solenoids for individual rows of bags,
(Courtesy of PEI Associates, Inc.)
4-25
-------
oil in the pulse-jet compressed-air supply. Compressed-air systems can be
equipped with small water and oil traps that work well if the compressor is
maintained and the humidity is not excessive. Even these systems, however,
must be drained periodically to be effective. Water and/or oil that are
blown into the bags during cleaning tend to absorb through the bag and cause
bag blinding as the dust cake becomes wet. The result is usually excessively
high pressure drop through blinded bags, which must be thrown away. The oil
usually comes from leakage around worn rings and seals in the compressor and
the moisture comes from the atmosphere. The rate at which the bags are
affected depends on how much water and oil enter the system.
4.3.6 Problem Identification
The key to effective fabric filter performance is a good design that
facilitates proper maintenance, a good understanding on the part of mainte-
nance personnel as to how the equipment is supposed to work and what can go
wrong, and the existence of a diagnosis and corrective action plan aimed at
correcting the problem not the symptom. Many of the problems discussed in
this section produce several common results: they generally increase bag
pressure drop, they may shorten bag life, and they may result in higher
opacity from the fabric filter outlet. Trying to isolate the cause of the
problem is usually the most cost-effective approach because such action may
avoid costly bag replacement or repetitive failures that cause long-term
performance to suffer. Some synergistic effects are possible with the fabric
filter, but they are relatively limited in scope. Usually repetitive fail-
ures in a fabric filter indicate that the cause or causes have not been
properly identified. Sometimes one must check the actual equipment against
the original design to ascertain that the fabric is identical to that speci-
fied and that it is properly installed. When these questions have been
satisfactorily addressed, various operating problems should be investigated.
4.4 CORRECTIVE ACTIONS
Fabric filters generally have a potentially high collection efficiency,
but they must be maintained properly to achieve acceptable long-term perform-
ance. This includes preventive maintenance and the correct diagnosis and
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS 4-26
-------
solution to problems as they occur. Some of the corrective actions are
obvious. Even so, the bulk of the maintenance effort seems to be aimed at
correcting the symptoms of problems rather than finding the cause and cor-
recting it to avoid recurrence. For example, if a set of bags are destroyed
by burning, they obviously must be replaced. That solves the immediate
problem of high emission rates, but it doesn't answer the following ques-
tions. What caused the bag damage? Is there a high temperature alarm and
recorder in the system? Does it work? Could this have been prevented, and
can it be prevented in the future? Trying to find the cause of a problem and
to correct that cause involves a different approach and attitude than simply
treating the symptoms. This difference should be recognized by both plant
and regulatory personnel. This section stresses how this difference can
effect long-term compliance expectations and in some cases, control costs.
4.4.1 Fabric Failures
Installation--
Failure to install filter bags properly almost certainly guarantees
future problems with the bags. For example, on pulse-jet fabric filters
aligning and sealing the bag at the tubesheet are somewhat difficult tasks.
In some applications (e.g., cement clinkers), the dust is very abrasive and
will eventually wear away the tubesheet at points where the bag is not sealed
properly. This can necessitate replacement of the tubesheet if the wear is
so significant that the bags can no longer be sealed, even with extra effort
and attention on the part of maintenance personnel. When this occurs, the
lost production and equipment replacement costs are substantial as a result
of a problem that could have been avoided had the bags been installed prop-
erly at the outset. Figure 4-9 demonstrates correct and incorrect methods of
installing bags in a reverse air fabric filter.
A little more attention to detail and proper installation can often make
the difference between good and poor performance and acceptable and unaccept-
able bag life.
High Temperature—
The effects of high temperature conditions can range from a few holes in
the bags caused by sparks to complete destruction of all the bags resulting
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS
4-27
-------
CORRECT AND INCORRECT
BAG INSTALLATION
oo
Incorrect
Example of correct and incorrect methods of installing bags in a
reverse air fabric filter.
(Courtesy of Wheelabrator-Frye, Inc.)
-pi
oo
-------
from generally high temperatures within the fabric filter. Although damage
from high temperatures is usually limited to sources operating at elevated
temperatures, it cannot be ruled out for sources where spark carryover into
the fabric filter is a potential problem. For example, in the furniture and
woodworking industry, fabric filters applied to sanders and abrasive planers
have the potential to throw a spark into the gas stream that may then be
carried to the fabric filter. The ensuing fire or explosion will destroy the
bags just as surely as if the air temperature had been raised over the bag
temperature limitation. The use of temperature monitors is usually recom-
mended at sources that operate at elevated temperatures, and plant personnel
generally install alarms and perhaps an emergency bypass system to protect
the system from temperature excursions. A temperature monitor would be
useless in the example cited above, however, because the temperature sensor
would not react quickly enough to take action to avoid the situation. Spark
arresters have been used with some success where sparks (not high gas temper-
ature) have< proven to be a problem.
When high temperature damage to the bags does occur, the cause of the
temperature excursion (e.g., operator error, process upset and nature of
upset) should be determined and action taken to prevent the occurrence of the
problem. This might entail the installation of temperature recorders where
none existed previously, the addition of temperature-conditioning systems or
an emergency bypass, education of operators to avoid certain conditions, or
combinations of conditions that may lead to high temperature excursions.
Proper identification and correction of the cause of high temperatures
usually prove to be much less expensive than periodic replacement of the
bags.
Condensation—
The condensation problem can generally be corrected by increasing
operating temperature, decreasing moisture and/or acid gas levels entering
the fabric filter, or by insulating the fabric filter more effectively (if
insulation already exists). Removal of the gases prior to shutdown and
preheating the fabric filter before startup (usual purge and preheat times
range from 5 to 20 minutes) also may minimize the potential for condensation
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-29
-------
within the fabric filter. All of these points could be considered as changes
in operating practice.
When a fabric filter already has a bag blinding problem as a result of
condensation, little can be done but change the bags and try to avoid a
recurrence of the conditions that caused the problem. In some situations,
allowing the bags to "dry out" by passing hot, dry gas through the system,
may enable the cleaning system to remove enough material for the fabric
filter to become operational again. Some permanent increase in pressure drop
is likely to remain in this situation, however, which means increased energy
cost to the plant and potentially shortened bag life.
Another option that has been used with some success is to remove and
wash the bags and then place them back in service. The cost varies according
to bag size and construction, but it generally runs approximately half that
of the cost for new bags. The bags are removed from the fabric filter and
checked individually and the integrity of each is checked. Only those bags
that appear to be in good condition are washed or dry-cleaned; other bags are
replaced. Before being placed back in service, the cleaned bags are again
screened to check for fabric integrity. Those bags that do not pass must
also be replaced. For bag washing to be effective, only a small percentage
of bags (i.e., less than 10%) can be rejected during the screening process
and at least half of the expected bag life must remain. Otherwise, bag
washing or dry cleaning does not appear to be a cost-effective approach.
When the particulate matter to be captured by the fabric filter is
expected to be sticky or condensible material, a precoating may be injected
into the gas stream to coat the bags and keep this sticky material from
condensing on or in the bags and blinding them. Any dry, powdered inert
material may be used, but pulverized limestone is the most common. This
material must be added continually to protect the bags. When limestone is
used, care must be taken to prevent temperatures from falling below the
moisture dewpoint. Otherwise, the limestone will "set" and blind the bags.
Fly ash and other materials have also been used for precoating the bags.
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-30
-------
Chemical Degradation--
Bags damaged by chemical degradation generally must be replaced. Once
begun, loss of fabric strength due to chemical degradation cannot be re-
versed. After the damaged bags are removed, consideration should be given to
altering the temperature (if condensation or increased degradation is occur-
ring at current conditions), to removing or reducing the offending constit-
uent, or to changing to a less susceptible fabric. If chemical degradation
is occurring near cool surfaces in the fabric filter, improving the insula-
tion or installing windbreaks may help to correct localized chemical degrada-
tion problems.
High A/C Ratio—
When high A/C ratios are known to be a problem, the two most viable
solutions usually are to reduce the gas volume through the system or to
increase the filter area by installing additional fabric filtration capabil-
ity. The costs of operating a system at a higher than normal design range of
A/C ratios are generally related to the energy required to move the gas
through an increased resistance (pressure drop), and the increased bag re-
placement costs resulting from bag abrasion, blinding, and generally short-
ened bag life. In some situations, the cost of operating and maintaining the
system at high A/C ratios far outweighs the cost of adding additional filter
area or finding ways to reduce the gas volumes through the fabric filter
system. This is particularly true when the fabric filter is applied to a
combustion source. Combustion and thermal efficiency are related to the
amount of excess air used in the combustion process. In a boiler, for
example, the thermal efficiency decreases as the percent excess air in-
creases. Thus, for a given boiler operating at a fixed steam production
rate, more fuel is required per pound of steam at higher excess air con-
ditions than at lower excess air conditions because of changes in thermal
efficiency. The net result of burning more fuel at higher excess air levels
is an increase in the quantity of flue gas produced and, therefore, an
increase in A/C ratio. This in turn produces a higher pressure drop across
the fabric filter and the potential for bag damage and shortened bag life.
Energy costs alone are usually substantial enough to merit changes in the
SECTION 4-PERFORMANCE EVALUATION. PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS
4-31
-------
operation of the unit, but personnel must look beyond the symptoms to the
causes and evaluate what can be done to improve performance and reduce costs
at the same time.
High Pressure Drop--
Because, as stated earlier, high pressure drop is usually a symptom of
some other problem, personnel must seek the cause of the problem within the
system.
Bag Abrasion--
Like high pressure drop, bag abrasion is often a symptom of a problem
elsewhere in the system. Causes of bag abrasion include bag-to-bag contact,
poor tensioning of bags, lack of or wear of a baffle plate or other preclean-
ing device, short or no thimbles (in some shaker and reverse-air systems),
and high pressure drop as a result of high A/C ratios or bag blinding. All
of these problems are related to installation, design, or operating problems,
most of which are discussed elsewhere in the manual.
If the fabric filter includes precleaning devices and blast plates,
these should be checked periodically for wear, as they can wear out quite
quickly and allow carryover of heavier, more-abrasive particles to the
fabric. When wear of the bags opposite from the inlet is consistent in
fabric filters not equipped with blast plates or long thimbles, equipping the
fabric filter with a method for protecting the bag may be worthwhile. For
bags mounted on thimbles, the thimbles should be well-rounded with no sharp
edges and the bags should be properly tensioned. Again, determining the
cause of the bag abrasion rather than just replacing the damaged bags may
ultimately save maintenance time and money.
4.4.2 Dust Discharge Failures
Hoppers should not be used for long-term storage. Actually, continuous
removal of dust from the hoppers is preferred to minimize compaction and
hopper bridging. At sources operating at elevated temperatures, this can be
particularly important because many dusts have better flow characteristics
when they are warm than when they are cold. To assist those persons assigned
to check the equipment periodically, markers should be placed on the shafts
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-32
-------
of the dust discharge system for easy confirmation that the equipment is
operating.
Dust discharge systems must maintain a seal in negative pressure ap-
plications. If the fabric filter is located ahead of the fan, the system
will most likely be under negative pressure. The hopper, dust discharge, and
airlock system should be free of air inleakage to reduce or eliminate the
resuspension of particles into the gas stream, and at sources operated at
elevated temperatures, to reduce the cooling effects attendant with inleak-
age. If hopper pluggage is due to cooling of the dust, eliminating inleak-
age, installing proper insulation, and using hopper heaters are possible
corrective alternatives. This assumes that the dust discharge is adequately
sized. If such is not the case or if particle characteristics change, the
system may be unable to handle the quantity of dust delivered.
Vibrators can sometimes cause further compaction of the dust in a hopper
rather than helping the dust flow, which should be considered in a decision
of whether to use a vibrator. The use of a sledge hammer by plant personnel
also should be approached carefully. In some situations, beating on the
hopper only compacts the dust and puts dents in the hopper that provide
future sites for further hopper pluggage. Taken to extreme, the hoppers can
become so distorted that they remain constantly plugged or holes form that
allow inleakage or fugitive emissions, depending on how the system is de-
signed.
In all cases, hoppers should be cleared as soon as possible to prevent
bag blinding, to avoid damage to the dust conveying system, and to minimize
the fugitive emissions generated by manually emptying the hoppers.
4.4.3 Shaker Cleaning System Failures
When a system or module is isolated for cleaning, there should be no
flow of gas through the system. Although damper activation can usually be
checked visually during operation, the integrity of sealing cannot. Depend-
ing on the damper and fan arrangement of the system, a pressure differential
across the module should show either a zero pressure drop or a static pres-
sure equal to the outlet value. If the value measured for the isolated
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS
4-33
-------
compartment varies from these values, some gas is still flowing through the
compartment. This can make the shaking action less effective in removing the
participate matter from the bags, and bag blinding and higher than normal
pressure drops may result. Although shaking action can be intensified
somewhat to counteract the less effective cleaning, such action could damage
the bags.
Shaker motors and shaker mechanisms should be kept in good operating
condition. They should be checked periodically as part of a preventive
maintenance plan, and any broken or worn parts should be replaced.
Bag tension and bag suspension should also be checked periodically. New
bags may stretch or shrink when exposed to the gas conditions in the fabric
filter. Approximately 2 weeks after initial operation, new bags should be
checked for proper tension and adjusted as necessary. It should be noted
that the tension on the bag will vary with dust cake loading. It is probably
better to shake all the bags so that some valid comparison of tension can be
made before adjustments are made. Records should be kept of the adjustments
made and the observed dust cake release to determine if any trends or pat-
terns are occurring.
4.4.4 Reverse-Air Cleaning System
Failures of the reverse air cleaning system are usually related to poor
bag tension, failure of isolation dampers, and failure of the reverse-air
fan(s). As in the shaker type fabric filter, isolation of the module being
cleaned is essential for proper operation. It is not the reverse flow of gas
that removes the dust from the bag, but the flexing action of the bag itself.
Failure of the cleaning system to work in a coordinated action tends to leave
an excessive dust cake on the bags, a higher pressure drop, greater bag
abrasion, and possibly reduced bag life.
Failure of the reverse-air fan would allow the bags to hang in the
fabric filter when the compartment is isolated, but no energy would be
applied to flex the bags, so excessive dust cake buildup eventually occurs.
The fan should be periodically checked for proper operation, and on some
larger systems, the reverse-air fan motor current should be monitored to
ensure proper operation.
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS
4-34
-------
Failure to seal the isolation dampers or open the reverse-air damper can
present problems. On most systems the damper drive systems (pistons, etc.)
can be visually checked. In many cases, however, confirmation that the
dampers are fully closed and sealed is not possible. If monitored, pressure
drop across the tubesheet during the "dwell" period should show 0 in. H20
because the gas flow through the compartment would cease. If the manometer
is working correctly and some pressure drop is observed, one could conclude
that the dampers were not sealing properly, and maintenance should be sched-
uled to adjust or repair the dampers.
Bag tension should be checked periodically, particularly after bag
replacement. Approximately two weeks after replacement, the tension should
be checked and adjusted as necessary. Because dust layer thickness may
affect bag tension, the tension on all bags should be compared after cleaning
so that proper tension is provided. Excessive buildup in the bags should be
noted, recorded, and evaluated for trends or patterns within the fabric
filter.
4.4.5 Pulse-Jet Cleaning System
As discussed earlier, problems with pulse-jet systems that involve bent
cages, leaks in bag/tubesheet seals, and bag-to-bag contacts primarily result
from improper installation. This difficulty can be eliminated through proper
training of personnel responsible for bag installation. Bag abrasion result-
ing from wear or lack of a baffle plate have also been discussed. The
corrective actions discussed in this subsection focus on the compressed-air
system.
The compressed-air pressure must fall within a specific range to gen-
erate a shock wave that traverses the bag length and returns, and thereby
flexes the dust cake and causes its removal from the bag. The pressure range
is partially related to the length of the bag; higher pressures are required
for longer bags. If pressure is insufficient, the bags will not be cleaned
properly and pressure drop will begin to rise. Insufficient pressure con-
ditions can result from an undersized compressor, a leak in the system, a
large number of systems being served by the compressor at one time, or
someone's closing a supply-line valve. If the compressor is inadequate to
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS. AND PROBLEM SOLUTIONS 4-35
-------
handle all the needs of the various systems, additional compressor capacity
may be required. The cost of extra compressor capacity would be offset by
lower energy costs resulting from the fabric filter operating with cleaner
bags. The various systems supplied by the compressor(s) also should be
checked to ensure that no leakage or otherwise wasteful use of compressed air
is occurring. For example, a leak pulse-pipe diaphragm allows compressed air
to escape and lowers compressed-air pressure.
Too high a pressure (above ^115 psig) creates a different problem. This
occurs when enough pressure is not available to clean the entire length of a
long bag. The pressure can be so high at the top of the bag that it blows
holes or causes tears in the fabric. A diffuser insert manufactured by
Sta-Clean is supposed to help equalize the pressure wave at the top and
bottom of the bag and even to allow operation at a lower compressed-air
pressure. Operating experience with this device seems to confirm that the
tops of the bags are protected and more uniform cleaning of the bags occurs.
This should extend bag life and lower energy requirements.
Bag blinding resulting from the presence of water and oil in the com-
pressed-air supply can be solved in several ways. First, routine maintenance
of the compressor can prevent worn compressor rings from passing oil into the
compressed air systems. Second, a trap and/or air in-line dryer can be used
to remove any water and oil. Third, the surge tank should be located such
that compressed air entering the pulse pipe exits the tank from the top
rather than the bottom. With this design, any water and oil that enter the
surge tank will tend not to leave the tank except through a blowdown valve
provided on the bottom of the tank. Figure 4-10 is an example of a top-
mounted surge tank that can allow moisture to enter the pulse pipe. Lastly,
if the compressor is beyond reasonable repair, a new one should be consid-
ered. Again, cost is important in the consideration of whether to repair or
replace a compressor or install a dryer in the system. If the cost of one or
two bag changes, however, is equivalent to the cost of a new compressor and
the bags are lasting less than half of their normal expected life, the cost
of replacement will be offset by the extended bag life. Another set of costs
often overlooked are those associated with lost production caused by the
SECTION 4-PERFORMANCE EVALUATION. PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS 4-36
-------
Figure 4-10.
Diaphragm assembly for pulse-jet system with solenoid connection removed.
Top-mounted surge tank can allow oil and moisture
into blow pipe and subsequently onto the bags.
(Courtesy of PEI Associates, Inc.)
-------
fabric filter being down and the cost of maintenance personnel to change the
bags. The solution to the problem of bag blinding is to identify the cause
of bag blinding and make the appropriate changes. Cost savings can be
substantial. At one fabric filter installation, bags had to be changed once
every 2 months because of bag blinding at an approximate cost of $5000 for
bags alone. The cause of the problem proved to be a worn compressor that was
losing quite a bit of oil, and the cost of replacing the compressor with one
of equivalent size was approximately $5000. Even with a bag life of only 1
year (half of the expected normal bag life of 2 years), this company spent
$25,000/year more on bags than necessary, and if the cost of lost production
and maintenance personnel were added, the actual cost would be at least twice
that amount. As has been stressed several times, identifying the cause
rather than treating the symptom is usually the least-cost solution to any
problem.
In the pulse-jet activation circuit, leakage of the diaphragm can cause
compressed air to escape from the surge tank. This lowers overall pulse
pressure, reduces cleaning efficiency, and increases fabric filter pressure
drop. A continuous hiss from the leaking diaphragm is usually an indicator
of this condition. Most plant maintenance personnel keep several spare
diaphragm replacement kits available because repair is usually relatively
simple. The solenoids that activate the various pulse-pipes are also subject
to failure and are easily replaced in most designs. When these solenoids
fail, the diaphragm will not open and material is allowed to build up in the
row that is supposed to be cleaned. Depending on the fabric filter design,
this problem may not be detected until someone checks for the activation of
each row for cleaning.
The timing circuit is also subject to failure. The older fabric filters
used mechanically driven and activated rotary switches to activate each
solenoid. Newer designs use electronic timers to activate the cleaning
system. These new systems tend to be more compact and more reliable if
maintained properly. Both systems require clean, dry mountings to operate
properly, and the presence of water and dust inside the timer enclosure can
lead to failure of the timer circuits. The electronic timers also must be
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS 4-38
-------
properly shock-mounted in a cool area. Electronic timers mounted on vibrat-
ing machinery have occasionally suffered from cracked circuit boards or
loosened components as a result of the vibration. In addition, the solid-
state components generally cannot withstand temperatures above 115° to 125°F
for extended periods of time. Failure of the timing circuit will cause the
bags to blind because they cannot be cleaned.
The last problem to be discussed with regard to pulse-jet systems is
pulse-pipe alignment. When installed or replaced, the pulse pipes should be
aligned over the row of bags so that the openings are centered over each bag
and aimed down the center!ine of the bags. The attachment technique on many
pulse-pipe designs ensures this alignment. In some pipe designs, however,
the pipes can be misaligned, as shown in Figure 4-11, and care should be
taken to be sure these pipes are properly aligned. In addition, the end of
the pipe that is away from the pulse supply must be bolted or clamped down.
If this clamp is broken, pipe misalignment is likely and damage to the top of
the bags can occur. Pipes that are loose usually create a rattle inside the
clean-air plenum when the cleaning system is activated. Maintenance should
be scheduled as soon as possible to correct this problem, as pulse pipes can
break from the connectors and damage the bags.
SECTION 4-PERFORMANCE EVALUATION, PROBLEM DIAGNOSIS, AND PROBLEM SOLUTIONS 4-39
-------
Figure 4-11. Misaligned blow pipe in pulse-jet fabric filter
caused by a broken bolt at the end of the pipe. (This results in this row not being cleaned.)
(Courtesy of PEI Associates, Inc.)
•t*
o
-------
SECTION 5
O&M PRACTICES
The importance of proper design considerations, timely detection of mal-
functions and bag failures, and good recordkeeping practices to fabric filter
performance has been discussed. Also essential to satisfactory long-term per-
formance of this control device are proper operating procedures and preventive
maintenance practices. Although high-temperature fabric filters often receive
the most attention, proper operating procedures are essential for both high-
and low-temperature applications. This section discusses general operating
procedures and preventive maintenance practices that can minimize unexpected
malfunctions and improve the performance of the fabric filter.
5.1 OPERATING PROCEDURES
Proper operating procedures are important during startup, shutdown, nor-
mal operations, and emergency conditions. For any system, and particularly a
new system, these procedures should include training of O&M personnel in
design fundamentals, component operations, and the limitations and expected
range of values of various operating parameters. Often this training is over-
looked, thought to be too expensive, or left to on-the-job training. Whereas
the cost of repairing the damage resulting from improper operation of some
systems is not excessive, the cost can be substantial for other systems (e.g.,
large high-temperature units), particularly if failures recur as a result of
continual operation in a manner that is incompatible with the system's design.
5.1.1 Startup Procedures
Improper startup can adversely affect many fabric filters. Most often
these adverse effects will be reflected as an increase in pressure drop or a
Shortening of bag life, both of which increase the cost of operating the
fabric filter.
SECTION 5-O&M PRACTICES
*J 'm J.
-------
For new fabric filters, a complete check of all the components is recom-
mended prior to operation. Any necessary repairs or corrections are usually
easier to make while the fabric filter is still clean. This component check
should include operating the cleaning system, the dust-discharge system, and
the isolation dampers and fans. It should also include passing clean ambient
air through the system to confirm that all bags are properly installed. As a
final check, fluorescent dye may be injected into the system to check for
proper sealing of the tubesheet and bags. The use of fluorescent dye is
usually reserved until some small amount of dust cake is built up on the bags
so that the dust will not bleed through the bags.
New bags are prone to abrasion if subjected to high dust loadings and
full-load gas flows. This is of particular concern during the initial start-
up, as new bags do not have the benefit of a dust buildup cake to protect the
fibers from abrasion or to increase their resistance to gas flow. Introducing
a full gas flow at high dust loadings can allow the particulate matter to
impinge on the fabric at high velocity and result in abrasion that may shorten
bag life. In addition, the dust may penetrate so deeply into the fabric that
the cleaning system cannot remove it, and a "permanent" pressure drop results.
Two methods are available to prevent this problem with new bags. The
first involves introducing a reduced gas volume into the fabric filter at a
lower mass loading (if possible) to allow the dust cake to build gradually and
gently. This prevents the particles from impacting the new bag fibers at full
velocity. The second method involves the use of a precoat material to provide
a protective filter cake before the process gas stream is introduced (see
Figure 5-1). The precoat material may be the same dust that will be filtered
during normal operation or some other dust that will provide suitable cake-re-
lease properties. Examples of precoat materials include fly ash and pulver-
ized limestone. Although the use of a precoat material may be part of normal
operation or routine startup procedures, it is also recommended when new bags
are installed, when an abrasive dust situation exists, when the bag fabric has
a low abrasion resistance, and when partial bag changeouts occur.
Dewpoint (both moisture and acid, if applicable) is a major concern
during startup. The presence of moisture in the gas stream usually does not
present a problem as long as the moisture is not allowed to condense within
SECTION 5-O&M PRACTICES
5-2
-------
PRECOATING MATERIAL FOR
PROTECTION OF BAGS FROM BLINDING
(Collection Inside Bag)
Cloth
Precoating
Material
Dirt
Smoke
Tar
tn
CO
Figure 5-1. Precoating material for protection of bags from blinding,
(Courtesy of the Carborundum Co.)
-------
the fabric filter. The introduction of warm moist gas into a cool or cold
fabric filter can cause condensation on the bags or on the fabric filter
shell. This moisture may cause bag blinding and a situation referred to
"mudded" bags. Although some dusts can be heated, dried, and then removed
from the bags, most will remain as a solid and perhaps impermeable dust cake
that produces excessively high pressure drops. Special care must be taken
with the use of some precoat materials (such as limestone) because these
materials tend to solidify when allowed to become moist (see Figure 5-2).
Preheating the fabric filter to a sufficiently high temperature to prevent
condensation is a practical alternative in some cases (e.g., fabric filters
used in asphalt plants can be preheated by firing the dryer without aggregate
until the temperature at the fabric filter outlet exceeds the dewpoint). In
the case of compartmented fabric filters, individual compartments can be
preheated and brought on-stream as the process rate increases.
Acid dewpoint may be important on some combustion processes (most notably
those using sulfur-bearing fuels). The acid dewpoint depends on the amount of
moisture and acidic material in the gas stream. Acid dewpoint conditions can
lead to corrosion of the fabric filter components, sticky particulate and
cake-release problems, and acid attack on some fabrics. One of the most
well-known but sometimes overlooked combinations that result in fabric acid
attack is the use of Nomex fabric with a gas stream high in sulfuric acid.
The sulfuric acid will attack the aramid structure of the Nomex fabric and
cause a loss of bag strength and fabric failure. If operators are not made
aware of the need to avoid acid dewpoint conditions, bag failure will occur
frequently.
Another problem (often associated with combustion sources) is unstable
combustion during startup. Poor combustion can produce substantial carbon
carryover, which may result in a sticky particulate. This situation also can
create the potential for fires in the fabric filter when a combustion source
and an adequate supply of oxygen are available. Because fires on the bags and
in the hoppers tend to destroy the fabric and necessitate bag replacement,
hoppers should be emptied continually during startup. Coal-fired boilers in
which oil is used during startup present problems in the areas of minimizing
emissions and protecting the fabric filter. Bypassing the fabric filter until
SECTION 5-O&M PRACTICES 5-4
-------
Figure 5-2. Limestone precoat hardened because of condensation
(dewpoint) problems. (Bags became blinded and high pressure drops resulted.)
(Courtesy of PEI Associates, Inc.)
5-5
-------
stable coal operation is achieved, preheating the fabric filter, and bringing
only the minimum number of compartments needed into service when stable com-
bustion has been achieved are all ways of avoiding operating problems. Con-
tinuous precoating of the bags can help to prevent sticky (condensable) soot
from fouling the bags during startup.
5.1.2 Normal Operating Procedures
During normal process operations, a well-designed and maintained fabric
filter should provide satisfactory control, which will be evidenced by its
outlet opacity. With the exception of condensable materials (water vapor,
heavy organics), most fabric filters should generate little or no visible
emissions. Opacity readings can help to determine the presence of pinholes
and tears in bags and, in some cases, the general location of the bag failure.
Whether determined by an opacity monitor or by a visual method, visible emis-
sions are usually the first indicator of poor fabric filter performance.
Pressure drop across the fabric filter also should be monitored periodi-
cally to assure that it remains within the expected range. Normally, the
equipment will assume some range of values (e.g., 2 to 6 in. I-LO), depending
on the dust loading, air-to-cloth ratio, and cleaning cycle. Inadequate
cleaning of the fabric, bag blinding, or excessive gas volume through the
system is generally reflected in the pressure drop. In some applications, the
pressure drop increases steadily as the bags age, and the "permanent" pressure
drop after cleaning also increases. Although many factors influence pressure
drop and bag life, pressure drop is still an extremely useful performance
indicator.
When used in conjunction with the pressure drop across the fabric filter,
measurement of fan motor amperage can also provide an indication of the
quantity of gas flowing through smaller fabric filters. In general, an
increase in current combined with an increase in pressure drop indicates an
increase in gas volume, and a decrease in amperage reflects a decrease in gas
volume. These changes, however, must be normalized for temperature (density)
changes because temperature influences the energy required to move the gas
through the system.
SECTION 5-O&M PRACTICES 5-6
-------
High-temperature operations should be equipped with continuous strip
chart recorders and high temperature alarms (see Figure 5-3). The high-
temperature alarms should provide some margin for corrective action, i.e., set
points of 50° to 75°F below the high temperature limit of the fabric. The
temperature alarm/recorder also may be connected to some automatic damper
system to control the temperature or to bypass the fabric filter. Although
some differential between the maximum temperature and the alarm activation
must be provided, the temperature set point should not be so low that the
alarm is continually activated. The temperature indicator will also monitor
against excessively low temperatures and dewpoint problems.
5.1.3 Shutdown Procedures
Dewpoint conditions and dust removal from the fabric filter are the pri-
mary concerns during shutdown. Failure to follow recommended shutdown proce-
dures also can result in early bag failure. Avoiding dewpoint conditions
through system purging is of top priority; bag cleaning and hopper emptying
are lower-priority items.
When processes operate on a daily cycle, the last operation of the day
should be to purge moisture and acidic materials from the fabric filter with-
out passing through the dewpoint. For example, an asphalt plant might allow
the dryer to operate for several minutes with the burner on after the aggre-
gate has been removed from the drum to remove moisture from the fabric filter.
Ambient air could then be drawn through the system to purge the remaining
combustion products from the fabric filter. Even well-insulated fabric fil-
ters usually have trouble maintaining temperatures above dewpoint for more
than several hours; therefore, it is advisable to purge these systems when
long idle periods are expected.
Upon shutdown, at least one or two complete cleaning cycles should be
allowed in compartmented fabric filters, and 5 to 20 minutes of cleaning in
pulse-jet systems. Removing the dust from the bags in this fashion will help
prevent blinding of the bags. Continuing to operate the hopper discharge
system while the cleaning system is in operation will minimize the chance of
hopper pluggage.
When emergency shutdowns of the fabric filter are necessary because of
high temperatures, spark detection, or other process upsets, the fabric filter
SECTION 5-O&M PRACTICES 5-7
-------
Figure 5-3. Examples of strip chart output on a fabric filter
with high temperature excursions as indicated by arrows.
(Courtesy of PEI Associates, Inc.) 5.3
-------
is usually bypassed to prevent failures and to protect the system from damage.
For such major problems as fires in the hoppers or on the bags, however, it is
probably better to let them burn out rather than to cut off the gas flow imme-
diately. Allowing the ignition source into a fabric filter without any gas
flow may cause an explosion. Also, adding water to a burning fabric filter or
to a hopper fire is not always advisable. In some situations, the addition of
water under reduced (oxygen-starved) atmospheric conditions will hydrolyze the
water and form hydrogen, which can create the potential for an explosion with-
in the fabric filter. The fabric filter manufacturer and insurance carrier
should be contacted whenever a known potential for fires/explosions exists.
Other process failures may necessitate only temporary bypassing of the
fabric filter, and the operation can be restored in a matter of minutes. In
these cases, the fabric filter generally does not have to be shut down com-
pletely and purged. If the upset cannot be corrected within a reasonable
amount of time, however, shutdown and the subsequent startup of the fabric
filter may then be necessary to prevent dewpoint problems.
It is important to note that bypassing the fabric filter during startup,
soot blowing, or an emergency may not be acceptable to the applicable regula-
tory agency. Such occurrences should be investigated and accounted for during
the design stages of development.
5.2 PREVENTIVE MAINTENANCE PRACTICES
The wide range of fabric filter applications makes specification of
preventive maintenance practices a difficult task. Recordkeeping is the heart
of any preventive maintenance program because it permits determination of pat-
terns that point to the possibility of major problems on the horizon. For a
fabric filter, recordkeeping centers on bag life and bag replacement, but
other items also must be considered.
5.2.1 Overall Maintenance Inspection Checklist
The following is a general checklist of items that should be inspected
regularly as part of a comprehensive O&M inspection program.
0 Inspect filter media for blinding, leakage, wear, slack, bag ten-
sion, and loose bag clamps, or discoloration.
SECTION 5-O&M PRACTICES
5-9
-------
0 Inspect the overall collector and compartment housings, hooding, and
connecting ductwork for leakage, corrosion, or dust accumulation.
0 Inspect all solenoid-operated pneumatic damper actuators, airlocks,
and valves for proper seating, dust accumulation, leakage, synchro-
nization, and operation.
0 Inspect hopper discharge for possible bridging of dust.
0 Measure the bag pressure drop. Compare frequency of cleaning with
that recommended by the manufacturer.
0 Inspect fan bolts (for tightness), bearings (for vibration), and
temperature. Inspect for erosion or dust buildup in the housing and
on the wheel. Check alignment of fan impeller with V-belt drive or
coupling and driver. Check sheave for signs of V-belt wear.
0 Inspect all bearings on fans, motors, dampers, etc., for lubrication
and free rotation.
0 Inspect foundation bolts on collector, motor, fan, etc., for tight-
ness. Also inspect bolts on collector housing and structural mem-
bers.
0 Inspect access doors(s) for leaks due to faulty gaskets or warping
of door(s) and/or frame(s).
Although the inspection frequency for an individual fabric filter system
will depend on the type of system and the vendor's recommendations, certain
major components should be inspected on a routine basis, and any needed main-
tenance should be performed. Table 5-1 summarizes the inspection and mainte-
nance schedule for the major components of a fabric filter system.
5.2.2 Daily Inspection/Maintenance
At least twice per shift (and perhaps as often as every 2 hours), opacity
and pressure drop should be checked. Sudden changes in these values along
with those of temperature and gas volume, may indicate a problem. For exam-
ple, the failure or partial failure of the cleaning system generally will
cause a relatively rapid increase in pressure drop in most systems. Timely
identification, location, and correction of this problem can minimize opera-
ting problems and long-term effects on bag life. Although identification and
subsequent correction of relatively minor problems have little effect on
fabric life, some minor problems tend to turn into major failures. Thus, the
SECTION 5-O&M PRACTICES
5-10
-------
TABLE 5-1. TYPICAL MAINTENANCE
INSPECTION SCHEDULE FOR A FABRIC FILTER SYSTEM1'6
Inspection
frequency
Component
Procedure
Daily
Stack and opacity monitor
Manometer
Compressed air system
Collector
Weekly
Damper valves
Rotating equipment and
drives
Dust removal system
Filter bags
Cleaning system
Hoppers
Check exhaust for visible dust.
Check and record fabric pressure
loss and fan static pressure.
Watch for trends.
Check for air leakage (low
pressure). Check valves.
Observe all indicators on con-
trol panel and listen to system
for properly operating sub-
systems.
Check all isolation, bypass, and
cleaning damper valves for
synchronization and proper
operation.
Check for signs of jamming,
leakage, broken parts, wear,
etc.
Check to ensure that dust is
being removed from the system.
Check for tears, holes, abra-
sion, proper fastening, bag
tension, dust accumulation on
surface or in creases and folds.
Check cleaning sequence and
cycle times for proper valve and
timer operation. Check com-
pressed air lines including
oilers and filters. Inspect
shaker mechanisms for proper
operation.
Check for bridging or plugging.
Inspect screw conveyor for proper
operation and lubrication.
(continued)
SECTION 5-O&M PRACTICES
5-11
-------
TABLE 5-1 (continued)
Inspection
frequency
Component
Procedure
Monthly
Quarterly
Shaker mechanism
Fan(s)
Monitor(s)
Inlet plenum
Access doors
Shaker mechanism
Semiannually
Annually
Motors, fans, etc.
Collector
Inspect for loose bolts.
Check for corrosion and material
buildup and check V-belt drives
and chains for tension and wear.
Check accuracy of all indicating
equipment.
Check baffle plate for wear; if
appreciable wear is evident,
replace. Check for dust
deposits.
Check all gaskets.
Tube type (tube hooks suspended
from a tubular assembly): In-
spect nylon bushings in shaker
bars and clevis (hanger) assem-
bly for wear.
Channel shakers (tube hooks
suspended from a channel bar
assembly): Inspect drill
bushings in tie bars, shaker
bars, and connecting rods for
wear.
Lubricate all electric motors,
speed reducers, exhaust and
reverse-air fans, and similar
equipment.
Check all bolts and welds. In-
spect entire collector thorough-
ly, clean, and touch up paint
where necessary.
SECTION 5-O&M PRACTICES
5-12
-------
operation of a fabric filter should be tracked on a daily basis to assure
early detection of any problems. The ability to perform on-line maintenance
depends on the design of the control equipment.
Routine checks of the fabric filter include pressure drop (and patterns
if a AP indicator and recorder are used), opacity patterns, dust discharge
operation, and external checks of the cleaning system operation. Other fac-
tors that can be checked include temperature (range) and fan motor current.
If a check of these factors reveals a sudden change, maintenance should be
scheduled as soon as possible.
5.2.3 Weekly Maintenance/Inspection
The extent of the weekly maintenance program depends greatly on access
and design of the fabric filter. Where possible, quick visual inspections
should be conducted; however, not all systems or processes are amenable to
this type of review. A weekly lubrication schedule should be established for
most moving parts. Manometer lines should be blown clear, and temperature
monitors should be checked for proper operation.
Shaker-Type Fabric Filters—
The operation of isolation dampers should be checked along with the
operation of the shaker system. The intensity of shaking should be relatively
uniform throughout the compartment. Bag tension should be checked, and any
fallen bags should be noted and repaired. The presence of any dust deposits
on the clean side of the tubesheet also should be noted, as well as any holes
or leaks in the bags.
Reverse-Air Fabric Filters--
The operation and sealing of the isolation and reverse-air dampers should
be checked. Each compartment should be checked for proper bag tension during
reverse-air operation, and any fallen bags should be noted and repaired. The
presence of dust deposits on the clean side of the tubesheets should be noted
to determine if there are any holes and leaks in the bags and if the seals are
tight. Tubesheets should be cleaned periodically to keep deposits from build-
ing up around the bags.
SECTION 5-O&M PRACTICES
5-13
-------
Pulse-Jet Fabric Filters--
On the dirty side of the tubesheet, bags should be checked for relatively
thin and uniform exterior desposits. Bags also should be checked for bag-to-
beg contact (points of potential bag wear). On the clean side of the tube-
sheet, each row of bags should be examined for leakage or holes. Deposits on
the underside of the blowpipes and on the tubesheet may indicate a bag fail-
ure. The cleaning system should be activated (the inspector should use hear-
ing protection), and each row of bags should fire with a resounding "thud."
The blowpipes should remain secured, and there should be no evidence of oil or
water in the compressed air supply. The surge tank or oil/water separator
blowdown valve should be opened to drain any accumulated water. Misaligned
blowpipes should be adjusted to prevent damage to the upper portion of the
bag. The compressed air reservoir should be maintained at about 90 to
120 psi.
5.2.4 Monthly-Quarterly Maintenance and Inspection
Beyond weekly inspections, the requirements become very site-specific.
Clear-cut schedules cannot be established for such items as bag replacement
and general maintenance of the fabric filter. Some items, however, may war-
rant quarterly or monthly inspections, depending on site-specific factors.
Items to be checked include door gaskets and airlock integrity to prevent
excessive inleakage (both air and water) into the enclosure. Any defective
seals should be replaced. Baffles or blast plates should be checked for wear
and replaced as necessary, as abrasion can destroy the baffles. Some facil-
ities prefer to use fluorescent dye to check the integrity of the bags and bag
seals (see Figure 5-4). Any defective bags should be replaced, and leaking
seals should be corrected.
Bag failures tend to occur shortly after installation and near the end of
a bag's useful life. A record of bag failures and replacements is invaluable
for identifying recurrent problems and indicating when the end of bag life has
been reached. Initial bag failures usually occur because of installation
errors or bag manufacturing defects. When new bags are installed, a period
SECTION 5-O&M PRACTICES
5-14
-------
Figure 5-4. Use of an ultraviolet light to check for leaks of
fluorescent dye that has been injected into the fabric filter.
(Courtesy of BHA Baghouse Accessories Co.,
Division of Standard Havens, Inc.)
5-15
-------
with few or no bag failures is normally expected unless serious design or
operation problems exist. As the bags near the end of their useful life,
however, the number of bag failures may increase dramatically. When weighed
against factors such as downtime for rebagging, the cost of new bags, and the
risk of limited production as the result of keeping the old bags in service,
the most economical approach may be just to replace all the bags at one time
to eliminate or minimize failure rate.
In some cases, bags can be washed or drycleaned and reused, e.g., when
dewpoint limits are approached or the bags are blinded in some manner. This
is generally an economically viable option when more than half a bag's "nor-
mal" life expectancy remains. Although cleaning may shorten bag life some-
what, sometimes it is economically more feasible to clean the bags than to
replace them.
SECTION 5-O&M PRACTICES 5-16
-------
REFERENCES FOR SECTION 5
1. Zurn Industries. Start-Up Procedure. Operation and Proposed Maintenance
Instructions. Air Systems Division, Birmingham, Alabama. 1980.
2. Flex-Kleen/Research Cottrell. Installation, Operating and Maintenance
Manual. Bulletin 399. Chicago, Illinois. 1980.
3. Ecolaire Environmental Co. Operation and Maintenance Literature.
Pleasant Hill, California. 1980.
4. W. W. Sly Manufacturing Co. Instruction Book, No. 693. Cleveland, Ohio.
1980.
5. Mikro-Pul Corp. Owner's Manual Micro-Pulsaire Dry Dust Collector,
SD379a. Section VII C - Troubleshooting. Summit, New Jersey. 1980.
6. Mikro-Pul Corp. Owner's Manual, Micro-Pulsaire Dry Dust Collector,
SD578, Section VII C - Troubleshooting. Summit, New Jersey. 1980.
SECTION 5-O&M PRACTICES 5-17
-------
SECTION 6
INSPECTION METHODS AND PROCEDURES
This section presents step-by-step procedures and techniques for de-
tailed external and internal inspections of fabric filters.
Fabric filter inspections are performed for the following purposes: as
part of system startup, for troubleshooting, to determine compliance with
regulations, and as part of an overall operation and maintenance (O&M) pro-
The purpose of any fabric filter inspection is to determine the current
operating status and to detect deviations that may reduce performance or
cause failure at some future date. For this reason, inspection programs must
be designed to derive maximum benefit from the information gathered during
the inspection.
A properly designed inspection program can be used for three purposes:
recordkeeping, preventive maintenance, and diagnostic analysis. Depending on
its purpose, the inspection may be conducted by operators, maintenance staff,
regulatory agency inspectors, or outside consultants (vendor representa-
tives).
6.1 PRESTARTUP INSPECTIONS
Because most fabric filters use a centrifugal fan, the compatibility of
the fan and the fabric filter is very important. On startup, the fan resis-
tance will be considerably lower than the operating design level because of
the new bags, and the gas delivery rate may exceed the design value. This
can have two undesirable effects: 1) the fan power level may rise to a point
where the motor will overload and/or the higher-than-design volume flow may
cause excess penetration; and 2) this higher volume flow may damage the
filter medium. ' ' The overall system resistance often is so high tiut
variations in the pressure between new and used bays are too small to have a
SECTION 6-INSPECTION METHODS AND PROCEDURES
6-1
-------
178
significant effect on fan capacity. ' ' On systems in which the design
pressure loss across the fabric represents a large fraction of the total
pressure loss, the potential overload situation can be minimized by damper
adjustment, selection of a fan with nonoverloading characteristics, or the
use of adjustable inlet vanes. All dampers and/or inlet vanes should be
partially closed during startup to reduce power consumption. After the
operating fan speed has been attained and the overall system temperature
approaches the operating level, the damper should be opened carefully to
Q
avoid motor overload.
Before the initiation of normal operation, the inspector should check
several items as indicated:2'3'5'6'9'10
0 Inspect all bag compartments and ductwork to see that joints are
tight. The general location of air leaks in either positive or
negative pressure baghouses can often be detected audibly, and
exact locations can be established by applying a soap solution to
the suspected leak area.
0 Inspect all bolts to ensure that they are tight. Inspect and
properly lubricate (as applicable) all threaded elements on clamps
and door latches for corrosion protection and easy access.
0 Inspect bags to ensure that they are secured to the floor thimbles
or cages. If bags are furnished with ground wires to guard against
sparking (and dust explosions), they should be securely connected
to the tube sheet, which must be well-grounded,
0 Inspect all system controls to verify installation and operation in
accordance with manufacturer's recommendations.9'U~1S
0 Inspect fan to ensure that it rotates in the proper direction. A
visual inspection is recommended to identify this occasional cause
of high pressure loss, reduced amperage, and diminished air hand-
ling capacity.
0 Inspect the fan and motor system for vibration, noise, and, in
particular, overheated bearings.
6.2 STARTUP INSPECTION PROCEDURES
The startup and shutdown procedures depend on the type of fabric filter
(shaker, reverse-air, or pulse-jet) and the process or emissions being con-
trolled. Two examples are used to illustrate the startup inspection proce-
dures. The first is a pulse-jet fabric filter and the second is a reverse-
SECTION 6-INSPECTION METHODS AND PROCEDURES
6-2
-------
12 14
air system, both used to control a combustion process. '
6.2.1 Pulse-Jet System
The following procedures should be followed during the startup of a
12
pulse-jet system:
0 Check to see that the inlet damper is opened fully after the bag
pressure differential has increased to 3 to 4 in. water.
0 Do not activate the timer controlling the compressed air pulses
until the differential pressure has reached 4 to 5 in. water unless
operating conditions require a lower pressure drop.
0 During normal startup with seasoned or conditioned bags, apply
power to all auxiliary equipment (except fan), energize the timer,
and start the compressed air system.
0 Turn on fan motor with the system damper nearly closed to prevent
motor overload during the starting power surge.
0 Maintain the pressure loss across the fabric within its preset
range by adjusting the pulse jet cleaning cycle. More frequent and
higher pressure pulses will reduce the bag pressure loss, whereas
the opposite actions will increase bag pressure loss should the
need arise to reduce dust penetration.
0 When the system is to be shut down, first turn off the fan and then
close the inlet and exhaust dampers. After waiting 15 to 30 minutes,
shut off the compressed air and timing circuit, along with any
auxiliary equipment.
0 Be sure the hopper(s) are emptied of material before turning off
the airlock and/or screw conveyor. This step will reduce the
chance of dust hangup and plugging due to compaction and sticking
of the dust in condensing atmospheres.
6.2.2 Reverse-Air System
The objective during startup of a reverse-air fabric filter is to pre-
vent flue gas from entering the fabric filter until the fabric filter is
completely preheated and the bags are precoated. By minimizing penetration
of fine particles into the fabric structure, these precautionary steps reduce
the chance of premature filter plugging or blinding. Prior to actual start-
up, the system should be checked to verify that all control elements, dam-
pers, and fi
procedures:
pers, and fans are functioning properly. The following are prestartup
SECTION 6-INSPECTION METHODS AND PROCEDURES
6-3
-------
0 Preheat compartments by using hopper heaters, if available.
0 Preheat inlet duct with flue gas from gas- or oil-fired boiler, if
available, and if the damper arrangements are such that this can be
accomplished.
0 Preheat fabric filter with hot flue gas from gas- or oil-fired
boiler, if available.
0 Precoat bag surfaces.
0 Visually inspect filtration surface to ensure that bags are coated.
6.3 ROUTINE PREVENTIVE MAINTENANCE INSPECTIONS
This section presents suggested procedures for performing routine preven-
tive maintenance inspections of typical pulse-jet, reverse-air, and shaker
type fabric filters.
6.3.1 Pulse-Jet Fabric Filters
Evaluation of Plume Characteristics—
An average opacity should be predetermined. Most pulse-jet collectors
operate with less than 5 percent opacity, so values approaching 5 percent may
suggest operating problems. If puffs are observed, the timing should be
noted so that it is possible to identify the row being cleaned just before
the puff.
Filtration System—
The pressure drop across the collector should be noted. If there is a
gauge, proper operation of the gauge should first be confirmed by observing
meter response during the pulsing cycle. If there is some question about the
condition of the gauge or its connecting lines, one line at a time should be
disconnected to identify any plugged or crimped lines (disconnecting lines
may not be possible if there is a differential pressure transducer connected
to the gauge lines).
If a properly operating gauge is not available, the static pressure drop
should be measured with portable instruments. These measurements should be
made at isolated ports installed specifically for the use of portable instru-
mentation. It is important to make the measurements on the inlet and the
outlet one at a time so that plugged tap holes and lines can be identified.
SECTION 6-INSPECTION METHODS AND PROCEDURES
6-4
-------
The operation of the cleaning system should be checked by noting the air
reservoir pressure. The ends of the reservoir and the connections to each of
the diaphragm valves should be checked for air leakage. Because these valves
are normally activated on a frequent basis, it is usually possible to observe
a complete cleaning cycle. Each valve should generate a crisp thud when
activated. Valves that fail to activate or that produce a weak sound when
activated are usually not working properly (see Figure 6-1). If too many of
these valves are out-of-service, the air-to-cloth ratios are probably high,
which can cause excessive emissions through the baghouse or inadequate pol-
lutant capture. Even if all diaphragm valves are working properly, reduced
cleaning effectiveness can result from the low compressed-air pressures.
If the compressed-air pressures are too high, especially for units
designed with a high air-to-cloth ratio, the intense cleaning action could
result in some seepage of dust through the bag fabric immediately after
cleaning, when the bag is pushed into the support cage. This will cause a
momentary puff of 5 to 10 percent opacity.
Holes and tears can lead to puffs of 5 to 30 percent opacity during the
cleaning cycle. During the pulse, the material bridged over these areas is
removed and the particulate matter is allowed to leak through (see Figure
6-2). As soon as the pulse dissipates, material tends to bridge over the
holes again, and the area eventually heals. As the holes and tears increase
in size, the duration of the puff also increases. Continuous emissions
result when the holes and tears become too large to bridge over.
The discharge of solids from the filter hopper should be observed if
this can be done safely and conveniently. Solids are usually discharged on a
fairly continuous basis (following each pulsing of a row).
Compressed-Air System--
The compressed-air system should be inspected to determine whether it
contains any water or rust deposits that could cause the system to malfunc-
tion. One quick method of checking whether the system has water or rust
deposits is to carefully open the valve on the blowdown system and observe
whether any water or other material is being expelled through the valve.
Also, if the system has oil traps, the traps can be visually inspected to
determine if any water or other material is retained in the trap.
SECTION 6-INSPECTION METHODS AND PROCEDURES 5,5
-------
CLEANING VALVE PROBLEMS
Air Supply
Figure 6-1. Cleaning valve problems.
[Illustration reproduced from "The Maintenance of Exhaust Systems
in the Hot Mix Plant (IS-52A)" published by the
National Asphalt Pavement Association.]
6-6
-------
(Courtesy
6-7
-------
6.3.2 Reverse-Air and Shaker Fabric Filter^
Evaluation of Plume Characteristics—
An average opacity should be predetermined. Most reverse air and shaker
collectors operate with less than 5 percent opacity. Values approaching this
may suggest operating problems. A drop in opacity when a specific compart-
ment has been isolated for cleaning usually indicates holes or tears in bags
in that compartment. Shaker collectors often have opacity spikes immediately
following the cleaning cycle. Both conditions warrant further evaluation.
Filtration System--
The pressure drop across the collector should be noted. If there is
a gauge, its proper operation should first be confirmed. If there is some
question about the condition of the gauge or its connecting lines, one line
at a time should be disconnected to identify any plugged or crimped lines
(disconnecting lines may not be possible if there is a differential pressure
transducer connected to the gauge lines).
If a properly operating gauge is not available, the static pressure drop
should be measured with portable instruments. These measurements should be
made at isolated ports installed specifically for the use of portable instru-
ments. It is important to make the measurements on the inlet and the outlet
one at a time so that plugged tap holes and lines can be identified. Care
must be exercised while rodding out tap holes because on some designs it is
possible to poke a hole in the bag adjacent to the tap hole.
The pressure drop across each compartment should be determined during
the cleaning cycle. In shaker collectors, the pressure drop during the
cleaning of a compartment should be zero. Nonzero values indicate damper
leakage problems. In reverse-air collectors, backflow will cause a measur-
able pressure drop with a polarity opposite that of the filtering cycle. If
no gauge is available and the unit operates at an elevated gas temperature,
the gas temperature should be measured. This can be done at a point on the
inlet duct to the collector or at one of the tap holes (if direct access to
the interior of the collector is possible).
SECTION 6-INSPECTION METHODS AND PROCEDURES
6-8
-------
The rate of solids discharge should be checked, if this can be done
safely and conveniently. Solids are usually discharged only at the
beginning of the cleaning cycling in each compartment.
Air leakage through access hatches, solids discharge valves, hopper
flanges, and fan isolation sleeves should be checked by listening for the
sound of inrushing air.
6.4 DIAGNOSTIC INSPECTIONS
This section presents a suggested procedure for performing diagnostic
inspections of typical pulse-jet, reverse-air, and shaker type fabric filters
20
when certain problems arise.
6.4.1 Pulse-Jet Systems
High Opacity (continuous or puffs)--
On top-load type designs, the clean side of several compartments should
be checked if these can be safely isolated and if no pollutant capture prob-
lems will result at the source origin. Even slight dust deposits can be a
sign of major problems (most of the dust in the clean-side plenum is carried
out because of the relatively high gas velocities). Dust near one or more
bag outlets may suggest inadequate sealing on the tube sheet. Holes and
tears may disperse dust throughout the top side of the tube sheet and make it
difficult to identify the bag with the hole. Fluorescent dye may be used
later to identify the problem.
High Pressure Drop, High Opacity, or Process Fugitive Emissions--
For a top-access system, the possibility of fabric blinding can be
checked from the top access hatch. Oil and water in the compressed air line
are sometimes partially responsible for the blinding that takes part of the
fabric area out of service.
For conventional pulse-jet collectors, the possibility for blinding can
only be checked at the dirty-side access hatch. Figure 6-3 shows an example
of easily removable dust. A crusty cake is sometimes evidence of excessive
moisture or sticky deposits on the bags.
SECTION 6-INSPECTION METHODS AND PROCEDURES
6-9
-------
-------
Continuously High Opacity, Frequent Bag Failures (primarily at bottom)--
Both types of pulse-jet collectors can experience possible premature bag
failure at the bottom if the support cages are slightly warped and the bags
rub at the bottom. This can be checked from a dirty-side access hatch, or in
some cases from below as shown in Figure 6-4. Note: Only the operator
(using extreme caution) should open the hatches at the tops of hopper areas.
Hot solids can flow rapidly out of these hatches.
The bag failure charts for the fabric filter should be examined. If a
distinct spatial pattern is apparent, the damage may be due to abrasion
(inlet gas blasting, inlet swirling, or rubbing against internal supports).
The date of the bag removal and the elevation of the apparent damage (T-top,
M-middle, B-bottom) enable identification of many common modes of failure.
By using such charts, operators have been able to minimize both excess emis-
sion incidents and bag replacement cost. A rapid increase in the rate of
failure often suggests significant deterioration of fabric strength due to
chemical attack or high temperature excursions.
When bags are removed from service, a simple rip test should be per-
formed. If it is possible to rip the cloth by inserting a screw driver and
pulling, the bag damage probably was the result of chemical attack, high
temperature excursions, moisture attack, or routine fabric exhaustion. Most
fabrics damaged by abrasion-related problems cannot be ripped, even near the
site of the damage.
High Opacity and Distinct Pattern to Bag Holes and Tears--
Bag and cage assemblies should be carefully inspected on removal. Often
the point of bag failure is next to a sharp point on the support cage.
Premature failure may also be caused by cages that do not provide enough
support for the fabric.
If all the bags have failed at the top, the compressed-air nozzles may
be misaligned (see Figure 6-5). This can cause the pulse to be directed at a
narrow area at the top of the bag.
6.4.2 Reverse-Air and Shaker Type Fabric Filters
Suspected Air Leakage, Low Gas Temperature, or Low Pressure Drop--
The 0~ and C02 levels at the inlet and outlet of combustion source
fabric filters should be checked. The measurement point on the inlet must be
SECTION 6-INSPECTION METHODS AND PROCEDURES
-------
Figure 6-4. Bag-to-bag contact in a pulse-jet fabric filter
resulting from poor alignment of cages during installation.
(Courtesy or PEI Associates, Inc.)
cr>
i
-------
COMMON BAG PROBLEMS
External Wear Burned Holes External Wear
Inside Abrasion,
Top of Bag
Figure 6-5. Common bag problems with pulse-jet fabric filters, including abrasion
at the top of the bags caused by misalignment of compressed air nozzles.
(Courtesy of National Asphalt Pavement Association)
cr>
i
-------
between the solids discharge valve and the tube sheet, so that potential
inleakage at this point can also be taken into account. There should not be
more than a 1 percent rise in the 00 levels going from the inlet to the
outlet (e.g., 6% 02 in and 7% 02 out).
Continuously High Opacity (during most of operating period) or Pressure Drop
Much Greater or Lower than Basel irie--
The presence and nature of the clean-side deposits should be checked by
viewing conditions from the access hatch. Note that the compartment must be
isolated by the operator before attempting to do the internal inspection.
All safety procedures must be carefully followed prior to entry.
The presence of snap ring leakage is often indicated by enlarged craters
in the clean-side deposits around the poorly sealed bags. Holes and tears
can sometimes be located by the shape of dust deposits next to the holes (see
Figure 6-6). Poor bag tension is readily apparent from the access hatch.
Improper discharge of material from the bags can often be confirmed by noting
that the bags close to the hatch are full of material one or more diameters
up from the bottom (see Figure 6-7). Deposits on the bags should also be
noted.
Anything more than a trace of material on the clean-side tube sheet is
indicative of probable emissions from this compartment that are substantially
above the baseline levels.
If the bag failure charts show a distinct spatial pattern, the damage
may be due to abrasion (inlet gas blasting, inlet swirling, and/or rubbing
against internal supports). Including the date of the bag removal and the
elevation of the apparent damage (T-top, M-middle, B-bottom) makes it pos-
sible to identify many common modes of failure. Operators using such charts
have been able to minimize both excess emission incidents and bag replacement
cost. A rapid increase in the rate of failure often suggests significant
deterioration of fabric strength. A simple rip test should be performed on a
bag recently removed from service. If it is possible to rip the cloth by
inserting a screw driver and pulling, the bag damage was probably the result
of chemical attack, high temperature excursions, moisture attack, or routine
fabric exhaustion. Most fabrics damaged by abrasion-related problems cannot
be ripped, even near the site of the damage.
SECTION 6-INSPECTION METHODS AND PROCEDURES 6-14
-------
Figure 6-6
fmer'
dust from the pinhole leak and it too
(Courtesy of PEI Associates, Inc.)
fail.J
CT>
I
-------
Figure 6-7. Checking for excessive build-up and poorly tensioned
bags in a reverse-air fabric filter. This bag passed both tests
(Courtesy of PEI Associates, Inc.)
6-16
-------
The compressed air system should be inspected to ensure that it is
installed properly and that it has aftercoolers, automatic condensate traps,
4
and filters as necessary for proper operation. The inspection should deter-
mine whether there is any water or rust deposits in the compressed-air system
that would cause the system to malfunction. One quick method of checking
whether the system has water or rust deposits is to carefully open the valve
on the blowdown system and observe whether any water or other material is
being expelled through the valve. If the system has oil traps, the traps
also can be visually inspected to determine if any water or other material is
retained in the trap.
6.5 RECORDKEEPING
The key to a good inspection and maintenance program is recordkeeping.
A record of all inspections should be maintained in the form of inspection
reports. Bag failure records were discussed in section 4. A brief narrative
discussion of the major deficiencies that were discovered and a recommended
course of action should also be prepared and filed with the inspection report
form.
6.6 SUMMARY
To summarize the preceding discussions, the following major items should
be addressed during the inspection of any fabric filter system:
1) Dust Capture and Transport System
The inspector should check all movable or stationary hoods and evaluate
the capture velocities, dust accumulation, static pressure, condition of
cleanout traps, integrity of ductwork, fan wear, and leaks or fugitive dust
emissions.
2) Fabric Filter System
The following are the major elements that should be evaluated during the
4
inspection of the fabric filter system:
0 Parameter monitors — including opacity or broken bag detectors;
manometers for pressure drop across fabric, compartments, cr entire
SECTION 6-INSPECTION METHODS AND PROCEDURES
6-17
-------
collector; indicators for cleaning sequence, cycle time, compart-
ments off line, temperature, volume flow, air-to-cloth ratio,
moisture, pulse-jet header pressure, and reverse-air flow.
0 Baghouse exterior—cleaning system operation; cleaning method;
overall condition of exterior housing, including structural mem-
bers, access doors, and gaskets, reverse air fan operation, and
shaker mechanism. External inspection will reveal visual evidence
of corrosion; warping of panels; faulty or missing gaskets; loose
bolts; and noise, odor, or elevated temperatures, which are indi-
cators of worn bearings, overstressed fan belts, and electric motor
problems.
0 Baghouse interior (if deemed necessary and is feasible)--condition
of bags: tears, pinholes, and sagging (inadequate tension). A
sagging or slack bag can result in the bag folding over the bottom
thimble connection and creating a pocket in which accumulated dust
can rapidly abrade and tear the fabric. Slackness also prevents
effective cleaning action with both reverse-flow or mechanical
shaking systems. Dust seepage or bleeding and/or pinhole leaks are
evidenced by dust deposits on the clean side of the fabric. Stain-
ing and stiffening of the dirty fabric indicates excessive caking
caused by moisture condensation or chemical reactions. The latter
condition leads to fabric blinding and excessive pressure loss as
well as to fabric failure. More than a 1/4-inch dust layer on
floor plates or isolated piles of dust suggests excess seepage
and/or torn or missing bags. Inspection of the inlet plenum,
including bag interior, will reveal any excess dust buildup on bags
and distribution plates. As a "rule-of-thumb" for smaller baghous-
es, if the amount of dust on a bag after cleaning is more than
twice the weight of the new (unused) bag, insufficient cleaning is
indicated. The condition of solenoid valves, poppet valves, me-
chanical linkages, and bag clamps are also indicated.11
Figure 6-8 lists the major data elements that should be obtained and
evaluated in a routine diagnostic inspection of a pulse-jet fabric filter.
Figure 6-9 lists the major data elements that should be obtained and evalu-
ated in a routine and a diagnostic inspection of a reverse-air and shaker
fabric filter.
SECTION 6-INSPECTION METHODS AND PROCEDURES
6-18
-------
Routine Inspection Data
Stack
Fan
Average Opacity
Duration and Timing of Puffs
None
Fabric Filter Inlet and Outlet Gas Temperatures
Inlet and Outlet Static Pressures
Presence or Absence of Clean Side Deposits
Air Reservior Pressure
Audible Checks for Air Inleakage
Qualitative Solids Discharge Rate
Diagnostic Inspection Data
Stack
Fan
Fabric Filter
Average Opacity
Peak Opacity During Puffs
Duration and Timing of PUffs
Inlet Gas Temperature
Speed
Damper Position
Motor Current
Inlet Gas Temperature
Outlet Gas Temperature
Inlet Static Pressure
Outlet Static Pressure
Inlet 02 and C02 Content (Combustion Sources)
Outlet 0? and C02 Content (Combustion Sources)
Qualitative Solids Discharge Rate
Air Reservoir Pressure
Frequency of Cleaning
Presence or Absence of Clean Side Deposits
Audible Air Infiltration
Figure 6-8. Routine and diagnostic inspection data for pulse-jet
20
fabric filters.
SECTION 6-INSPECTION METHODS AND PROCEDURES
6-19
-------
Routine Inspection Data
Stack
Fabric Filter
Average Opacity
Opacity During the
compartment)
Cleaning Cycles (for each
Inlet and Outlet Static Pressures
Inlet Gas Temperature
Rate of Dust Discharge (Qualitative Evaluation)
Presence or Absence of Audible Air Infiltration
Presence or Absence of Clean Side Deposits
Ripping Strength of Discarded Bags
Baseline and Diagnostic Inspection Data
Stack
Fabric Filter
Stack Test
Fan
Average Opacity
Opacity During the Cleaning Cycles (for each
compartment)
Date of Compartment Rebagging
Inlet Static Pressure (Average)
Outlet Static Pressure (Average)
Minimum, Average, and Maximum Gas Inlet Temperatures
Average 02 and C02 Concentrations (Combustion
Sources Only)
Time to Complete a Cleaning Cycle of all
Compartments
Length of Shake Period
Length of Null Period
Bag Tension (Qualitative Evaluation)
Rate of Dust Discharge (Qualitative Evaluation)
Presence or Absence of Audible Air Infiltration
Presence or Absence of Clean Side Deposits
Emission Rate
Gas Flow Rate
Stack Temperature
02 and Co2 Content
Moisture Content
Fan Speed
Fan Motor Current
Gas Inlet and Outlet Temperatures
Damper Position
Figure 6-9. Routine and Diagnostic Inspection Data for reverse air
shaker fabric filters.20
SECTION 6-INSPECTION METHODS AND PROCEDURES
6-20
-------
REFERENCES FOR SECTION 6
1. Billings, C. E., and J. E. Wilder. Handbook of Fabric Filter
Technology, Vol. 1. GCA/Technology Division, Bedford, Massachusetts.
EPA-APTD 0690 (NTIS PB-200-648), December 1970.
2. Reigel, S. A., and G. D. Applewhite. Operation and Maintenance of
Fabric Filter Systems. In: Operation and Maintenance for Air
Particulate Control Equipment. Young, R. A. and F. L. Cross, eds. Ann
Arbor Science, Ann Arbor, Michigan, 1980.
3. Cross, F. L., and H. E. Hesketh, eds. Handbook for the Operation and
Maintenance of Air Pollution Control Equipment. Technomic Publishing
Co., Inc., Westport, Connecticut, 1975.
4. Roeck, D. R., and R. Dennis. Fabric Filter Inspection and Evaluation
Manual. Prepared for U.S. Environmental Protection Agency under
Contract No. 68-01-6316. 1980.
5. Cheremisinoff, P. N., and R. A. Young, eds. Air Pollution Control and
Design Handbook, Part 1. Marcel Dekker, Inc., New York, 1977.
6. McKenna, J. D., and G. P. Greiner. In: Air Pollution Control Equipment
- Selection, Design, Operation and Maintenance. Theodore, L. and A. J.
Buonicore, eds. Prentice Hall Inc., Englewood Cliffs, New Jersey, 1981.
7. Stern, A. C., ed. Air Pollution, Volume IV, Engineering Control of Air
Pollution. 3rd Ed., Parts A and C Academic Press, Inc., New York, San
Francisco, London, 1976.
8. Kraus, M, N. Baghouses: Selecting, Specifying and Testing Industrial
Dust Collectors. Chem. Eng., 86(9):133-142, 1979.
9. Industrial Gas Cleaning Institute (IGCI). Operation and Maintenance of
Fabric Collectors, Publication No. 53. Suite 304, 700 N. Fairfax
Street, Alexandria, Virginia.
10. Beachler, D. S., and M. Peterson. APTI Course SI:412 Baghouse Plan
Review, Student Guidebook. EPA No. 450/2-83-005, April 1982.
11. W. W. Sly Manufacturing Co. Instruction Book, No. 693, Post Office Box
5939, Cleveland, Ohio, 1980.
SECTION 6-INSPECTION METHODS AND PROCEDURES
-------
12. Mikro-Pul Corp. Owner's Manual Micro-Pulsaire Dry Dust Collector,
SD379a. Section VII C - Troubleshooting. 10 Chatham Road, Summit, New
Jersey, 1980.
13. Mikro-Pul Corp. Owner's Manual, Micro-Pulsaire Dry Dust Collector,
SD578, Section VII C Troubleshooting. 10 Chatham Road, Summit, New
Jersey, 1980.
14. Zurn Industries. Start-Up Procedure. Operation and Proposed
Maintenance Instructions. Air Systems Division, Post Office Box 2206,
Birmingham, Alabama, 1980.
15. Flex-Kleen/Research-Cottrell. Installation, Operating and Maintenance
Manual. Bulletin 399. 222 South Riverside Plaza, Chicago, Illinois,
1980.
16. Ecolaire Environmental Co. Operation and Maintenance Literature. 380
Civic Driver, Pleasant Hill, California, 1980.
17. Campbell, P. R. Maintaining Proper Tension. Power, 1_24(2):92, 1980.
18. Reigel, S. A. Fabric Filtration Systems, Design, Operation, and
Maintenance.
19. PEDCo Environmental, Inc. Development of Pilot Inspection System for
Virginia Air Pollution Control Board. Prepared for U.S. Environmental
Protection Agency under Contract No. 68-01-6310, Work Assignment No.28,
April 1983.
20. Engineering - Science. Field Inspection Notebook Prepared for the U.S.
Environmental Protection Agency under Contract No. 68-01-6312, Work
Assignment No. 62 and 69, October 1983.
SECTION 6-INSPECTION METHODS AND PROCEDURES 6_22
-------
SECTION 7
SAFETY
The safety of plant personnel and agency inspectors during all aspects
of fabric filter O&M is of ultimate importance. Areas of concern include
confined area entry (oxygen deficiency and toxic gases), hazardous materials
(dust, metals, etc.), chemical burns, eye injury, and normal industrial
safety concerns such as moving equipment, falls, etc. With regard to
fabric filters, many of these concerns occur simultaneously and in a con-
fined area, which presents the potential for serious injuries to personnel.
With proper planning, safety equipment, and established procedures, opera-
tion and maintenance and inspections can be performed safely without risk
of injury.
Many of the potential hazards and proper procedures for addressing
them are discussed in the following subsections. Further information on
confined area entry and manufacturers of safety appliances can be found in
specific vendor maintenance manuals on installed units and in Occupational
Safety and Health Administration (OSHA) and National Institute of Occupa-
tional Safety and Health (NIOSH) publications.
7.1 HOPPER ENTRY
Fabric filter hoppers present special safety hazards. It is recom-
mended that hopper doors be interlocked and that the hopper doors be opened
only after the unit has been shut down. For economic reasons, however,
many companies substitute the use of padlocks for the key-interlock system.
In principle, the use of padlocks is equally as safe if proper procedures
are followed. The tendency, however, is to remove the lock and open hopper
doors prematurely to quick-cool a unit or clear a hopper pluggage.
SECTION 7-SAFETY
7-1
-------
The danger in opening doors comes from the discharge of material im-
pounded in the hopper. Dust that has accumulated in valleys or corners may
break loose during entry into the hopper and fall on the inspector, causing
minor injury or, in some cases, serious injury or suffocation by burying
him or her. The doors on hoppers must be opened very carefully, and care
must be taken to ensure that no accumulation of collected dust is impounded
behind the inner door.
Entry into hoppers for purposes other than maintenance should be
avoided. Maintenance that can be conducted outside of the hopper should be
attempted first. If entry is necessary, the bags should be thoroughly
cleaned and then steps should be taken to dislodge and discharge dust from
the hopper before entry. This can be accomplished by mechanical vibration
(vibrators, hammers, etc.) or poking, prodding, or air lancing. Removal of
accumulated dust should never be attempted from inside the hopper. This
material may become dislodged and move en masse into the inlet or outlet
field hoppers and completely fill the hopper.
All hopper doors should be equipped with safety chains or double
latches to prevent complete opening upon release. This will slow the loss
of material or ash in the event of accidental opening of a full hopper.
Most hopper inner doors have design features that, if properly used,
will ensure that no door is opened when dust is impounded behind it.
First, a pipe coupling with a plug should be installed in the door; removal
of the plug allows visual verification of dust impoundment. Second, a
pressure-type latch should be used that allows a portion of the door seal
to be released and causes a gap between the door and sealing jam. This
partial release will allow accumulated dust to flow out and indicate a par-
tially full hopper without the possibility of the door being fully opened.
A normal practice is to discharge the hoppers fully before entry and
after each period of dust removal. Whether the hopper is full may be
determined by striking the door with a hammer. An empty hopper door will
resound with a ring; a full hopper will produce a dull thud.
A further warning in connection with hopper entry involves the use of
hand grips, foot holds, etc., inside the hopper. Because of the possibil-
ity of dust buildup on protruding objects, manufacturers have avoided the
SECTION 7-SAFETY
-------
use of handholds and footholds in hopper interiors. Thus, the steep valley
angles and dust layer create a potential for a fall and injury when enter-
ing the door. Temporary ladders and handholds may be installed and used
when needed. Because of the angles and small door openings, back injuries
are the most common (other than abrasions). Outside access equipment
(scaffolds, ladder, handholds, etc.) should be installed in a manner that
minimizes the awkward nature of hopper door entry. Also, if nuclear hopper
level detectors are used, the radiation source (beam) should be shielded
from the outside before the hopper is entered.
Hopper evacuation systems (screws, drag chains, agitators, etc.)
should not be operated when persons are inside the hopper area or in an
area from which they could fall into the hoppers. Dust accumulation that
is discharged into the hopper can be considered a live bottom with moving
equipment. The dust becomes fluid and creates treacherous footing. Scaf-
folds on which persons are standing may shift and float, and that person(s)
may become engulfed in the collected material.
7.2 CONFINED AREA ENTRY
A confined space is an enclosure in which dangerous air contamination
cannot be prevented or removed by natural ventilation through opening of
the space. Access to the enclosed area may be restricted such that it is
difficult for personnel to escape or be rescued. The most common examples
of a confined space are storage tanks, tank cars, or vats. Depressed areas
(e.g., trenches, sumps, wells) also may have poor ventilation and be consid-
ered a confined space. A fabric filter system falls under the general
definition of confined space, and as such, requires special procedures and
precautions with regard to entry.
Potential dangers of confined space fall into three categories:
oxygen deficiency, explosion, and exposure to toxic chemicals and agents.
Personnel entering the fabric filter for inspection or maintenance must
assess the risks and potential dangers in each category and follow specific
safety precautions.
SECTION 7-SAFETY
7-3
-------
7.2.1 Oxygen Deficiency
Oxygen deficiency is the most common hazard. Any gas generated in a
confined space displaces the atmosphere and reduces the oxygen content
below the normal value of 20.9 percent. Out-gassing of combustible gases
(methane, hLS, organic vapors, etc.) from collected particulate matter can
result in local pockets with reduced oxygen levels. Further, application
of the fabric filter to combustion sources (e.g., utility boilers, indus-
trial boilers, cement kilns, incinerators) produces an atmosphere that is
extremely low in oxygen (2 to 10 percent). Purging of the unit during
cooling does not always completely replace the flue gases with ambient air,
and local pockets may remain.
Reduction of oxygen pressure below normal conditions has increasingly
severe effects on a person and eventually leads to death. Oxygen levels of
less than 16.5 percent result in rapid disability and death. Table 7-1
shows the effects of reduced oxygen concentrations for various lengths of
time. Because of the subtle effects of oxygen deficiency, the average
person does not recognize the symptoms and may ignore the danger. By the
time the person does recognize the problem, he may no longer be able to
remove himself from the dangerous environment.
7.2.2 Explosion
Explosive atmospheres can be created in confined spaces by the evapo-
ration of volatile components or improper purging of the fabric filter when
the process is shut down. Three elements are necessary to initiate an
explosion: oxygen, a flammable gas, vapor or dust, and an ignition source.
A flammable atmosphere is defined as one in which a gas concentration is
between two extremes: 1) the lower explosive limit (LEL) and the upper
explosive limit (UEL). A mixture of gas and oxygen in a concentration
between these values can explode if a source of ignition is present.
Possible sources of ignition include cigarettes, matches, welding and
cutting torches, and grinding equipment. The best means of preventing
explosion is to dilute the flammable gas below the LEL by ventilation. It
is not safe to assume that a source of ignition can be eliminated and to
allow work to continue in a potentially explosive atmosphere.
SECTION 7-SAFETY
7-4
-------
TABLE 7-1. EFFECTS OF VARIOUS LEVELS
OF OXYGEN ON PERSONSa
Concentration,
percent
Duration
Effect1
20.9
19.5
16.5
12-16
10-14
6-10
Below 6
Indefinite
Not stated
Not stated
Seconds to minutes
Seconds to minutes
Seconds
Seconds
Usual oxygen content of air
Minimum oxygen content for oxygen
deficient atmospheres (OSHA
Standards)
Lowest limit of acceptable stan-
dards reported in literature for
entry without air-supplied
respirators
Increased pulse and respiration,
some coordination loss
Disturbed respiration, fatigue,
emotional upset
Nausea, vomiting, inability to move
freely, loss of consciousness
Convulsions, gasping respiration
followed by cessation of breath-
ing and cardiac stand-still
Data for correspondence of Robert A. Scala, Ph. D., REHD, Exxon Corporation,
March 26, 1974.
DEffects - Only trained individuals know the warning signals of a low oxygen
supply. The average person fails to recognize the danger until he is too
weak to rescue himself. Signs include an increased rate of respiration and
circulation that accelerate the onset of more profound effects including
loss of consciousness, irregular heart action, and muscular twitching. Un-
consciousness and death can be sudden.
SECTION 7-SAFETY
7-5
-------
Work in a confined area may release flammable gases, which can in-
crease in concentration. Constant ventilation should be provided to main-
tain the concentration below the LEL.
Because many vapors are heavier than air, pockets of flammable gases
may develop. An effective monitoring program checks concentrations at
multiple locations and times during the exposure period.
7.2.3 Exposure to Toxic Chemicals and Agents
Depending on the application of the fabric filter, collected dust may
contain toxic chemicals or harmful physical agents. These compounds may
exist in the system or be created as a result of operations in the confined
area. Inhalation, ingestion, or skin contact have adverse health effects.
Most agents have threshold limit doses below which harmful effects do not
occur. Exposure above these threshold doses can cause acute or chronic
symptoms, depending on the compound. A quantitative assessment of each
compound and the threshold dose levels must be made before anyone is allowed
to enter the fabric filter. Typical toxic chemicals or species in the
fabric filter environment can include arsenic, cadmium, beryllium, lead,
alkali, and acids. Table 7-2 lists the allowable concentrations for entry
into confined spaces for several compounds.
TABLE 7-2. ALLOWABLE CONCENTRATIONS FOR ENTRY INTO CONFINED SPACES
Agent3
Hydrocarbons
Oxygen
H2S
Carbon monoxide
so2
Without air-supply
equipment
1% LEL max.
19.5-23.5%
10 ppm max.
30 ppm max.
5 ppm max.
With air-supply
equipment
20% LEL max.
16.5% min.
300 ppm
200 ppm
500 ppm
No entry.
permitted
Above 20% LEL
Below 16.5%
>300 ppm
>200 ppm
>500 ppm
If other contaminants are present, an industrial hygienist should be con-
sulted for the appropriate allowable limits.
3Work may be performed in oxygen-free atmospheres if backup systems are
available, such as air-line respirators, self-contained breathing apparatus,
and an emergency oxygen escape pack.
SECTION 7-SAFETY
7-6
-------
As noted in Table 7-2, entry may be permitted within certain limita-
tions provided the person is equipped with appropriate approved respiratory
protection. An assessment of the hazard, concentration, permissible expo-
sure, and protective equipment must be made before anyone is allowed to
enter a confined space.
Each facility must establish a confined-space entry policy that in-
cludes recognition of the hazards, atmospheric testing and analysis, venti-
lation requirements, selection and use of protective equipment, training
and education of personnel, and administrative procedures.
An important component of the program is recognition of the potential
hazard, which requires complete knowledge of the industrial process and
wash area. A cursory examination cannot prevent serious deficiencies;
detailed analysis is therefore recommended.
The second policy component involves ambient air monitoring. An
initial certification of gaseous concentrations must be made before entry
is permitted. This certification must be made by a qualified safety officer
with properly calibrated and maintained equipment. In general, a permit to
enter (with a time limit) may be issued and displayed at the point of
entry. Assuming that oxygen and gas levels do not change with time can be
dangerous; an effective program should include periodic revaluations of
concentrations after initial entry.
Hazard Recognition--
Each worker should be trained in use of protective equipment, potential
hazards, early warning signs of exposure (symptoms), and rescue procedures
(first aid, CPR, etc.). It is most important that each person recognize
that multiple fatalities can occur if proper rescue procedures are not
followed. If a worker is affected within the confined area and cannot
remove himself, rescue personnel must not enter the area without complete
self-contained breathing equipment. If the first worker is affected by an
unknown agent, it is highly probable that rescue personnel will be similarly
affected unless they have the proper protective equipment._ Because the
causal agent is not known, maximum protection must be used during the
rescue attempt.
SECTION 7-SAFETY
-------
Atmospheric Testing and Analysis--
Gas monitoring usually is conducted to determine the percentage of
oxygen, the percentage of lower explosive limit (hexane/methane, heptane,
etc.), hydrocarbon concentration (in parts per million), and carbon monox-
ide levels (in parts per million). If hydrogen sulfide or other toxic
gases are suspected, additional analyses may be conducted with detection
tubes or continuous gas samplers. The use of continuous gas samplers with
an audible alarm is recommended. The initial measurements should be per-
formed according to the following suggested procedures.
1. Remaining outside, a gas tester should check the vessel's oxygen
content, explosivity, and toxic chemical concentration by first
sampling all entry ports and then use probes to sample inside the
space. Caution should be used when testing for combustible
gases, as many meters need an oxygen level close to ambient
levels to operate correctly. This is one reason that the space
should be purged and vented before testing. Voids, subenclosures,
and other areas where pockets of gas could collect should also be
tested.
2. When initial gas test results show that the space has sufficient
oxygen, the gas tester can enter the space and complete the
initial testing by examining areas inaccessible from outside the
shell. He/she should wear an air-supplied positive-pressure
respirator during these measurements. Special care should be
taken to test all breathing zone areas.
3. If the results of the initial tests show that a flammable atmo-
sphere still exists, additional purging and ventilation are
required to lower the concentration to 10 percent of the LEL
before entry may be permitted.
4. If testing shows an oxygen-deficient atmosphere or toxic concen-
trations, all personnel entering the space must use an appropri-
ate air-supplied respirator.
After the initial gas te3ting has been performed, dust, mists, fumes,
and any other chemical agents present should be evaluated by either an
industrial hygienist or a trained technician. The results will indicate if
additional control measures are necessary. Physical agents such as noise,
heat, and radiation also must be evaluated, and if any are present, the
appropriate control measures (e.g., ear protection or rotating employees),
should be instigated.
SECTION 7-SAFETY
7-8
-------
The specified respiratory protection should be based on the hazard
assessment, i.e., the type of contaminant, its concentration, and the
exposure time. The type of respiratory equipment required for each species
is specified by NIOSH.
Respirators include basic particle-removing devices (dust, aerosol,
mist, etc.), air-purifying respirators (gas, vapors, etc.), and air-supply-
ing respirators (air-line, self-contained).
7.3 WORKER PROTECTION
7.3.1 Eye Protection
Dust collected by fabric filters is very fine and usually contains a
high percentage of particles less than 5 mm. The particles may be sharp-
edged or crystalline in nature. Because all surfaces in the fabric filters
are coated with dust, which may be easily dislodged and suspended during
internal inspections, protection is necessary to prevent dust from entering
the eyes. Goggle type protection is generally not effective because of the
inability of the frames to form a tight seal against the worker's face.
Effective eye protection consists of full-face protection or a snorkling
mask.
Eyes also may be subjected to chemical damage as a result of the dust
composition or species condensed onto the dust particles. The most common
active agents are sulfuric acid on fly ash particles and alkali agents in
cement applications. Each plant should collect samples of fabric filter
dust and specify eyewash solutions suitable for removing or neutralizing
the active components. Table 7-3 summarizes the kinds of applications
where potential eye hazards may exist.
TABLE 7-3. APPLICATIONS PRESENTING POTENTIAL EYE HAZARDS
Application
Fly ash
Cement
Municipal incineration
Copper converter
Potential active species
Sulfuric acid
Alkali
(NaOH, Na?S04, K2S04> etc.)
Hydrochloric acid
Sulfuric acid
PH
Acid
Alkaline
Acid
Acid
SECTION 7-SAFETY
7-9
-------
7.3.2 Hearing Protection
The fabric filter housing is a large open area with metal walls that
tend to magnify and reflect sound energy. When inspectors are inside the
unit, proper hearing protection should be used to limit sound levels to
maximum permitted exposure. Many types of hearing protection devices
(cotton, premolded inserts, foam, ear muffs, etc.) are available; selection
depends on individual preference and expected sound levels.
Limits of worker exposure to noise are based on both duration of
exposures and sound levels (dBA). Permissible levels for intermittent
noise and nonimpulsive levels are presented in Tables 7-4 and 7-5.
7.3.3 Skin Irritation
Depending on its composition, the dust collected in the fabric filter
can be acidic, alkaline, hydroscopic, or abrasive. When it contacts the
skin, this dust can cause burns or irritation. Workers can limit skin
contact area and thus prevent potential irritation by wearing long-sleeved
shirts and gloves during internal inspections. Depending on temperature
conditions and activity levels, coveralls or other full covering may be
worn.
TABLE 7-4. MAXIMUM PERMISSIBLE SOUND LEVEL FOR INTERMITTENT NOISE9
(A-weighted sound level, dBA)
Number of occurrences per day
Total time/8 hours
8 hours
6 hours
4 hours
2 hours
1 hour
i hour
\ hour
8 minutes
4 minutes
2 minutes
i
89
90
91
93
96
100
104
108
113
123
3
89
92
94
98
102
105
109
114
125
7
89
95
98
102
106
109
115
125
15
89
97
101
105
109
114
124
35
89
97
103
108
114
125
75
89
94
101
113
125
^160
89
93
99
117
125
aSource: The Industrial Environment
1973, p. 327.
- Its Evaluation and Control. NIOSH,
SECTION 7-SAFETY
7-10
-------
TABLE 7-5. THRESHOLD LIMIT VALUES FOR NONIMPULSIVE NOISE (ACGIH)3
Duration, hours/day
8.00
6.00
4.00
3.00
2.00
1.50
1.00
0.75
0.50
0.25
Permissible sound level, dBA
90
92
95
97
100
102
105
107
110
115
Source: The Industrial Environment - Its Evaluation and Control. NIOSH,
1973, p. 327.
7.3.4 Thermal Stress
Thermal stress associated with inspections and maintenance of a fabric
filter and its components must be considered in defining the time required
for repairs. Because of the dusty, humid conditions and limited access,
thermal effects may be severe. Also, if limited time is available for
purging and cooling the unit, entry may have to be made under elevated
temperatures.
The thermal stress placed on the worker is a function of several
variables, such as air velocity, evaporation rate, humidity, temperature,
radiation, and metabolic rate (work). In effect, the stress is indicated
by the need to evaporate perspiration.
A Heat Stress Index developed by Belding and Hatch (1955)* incorpo-
rates environmental heat [radiation (R) and convection (C), and metabolic
(M)] into an expression of stress in terms of requirement for evaporation
of perspiration. Algebraically the function may be stated as foTlows:
M + R + C = E req.
Belding and Hatch. Index for Evaluating Heat Stress in Terms of Resulting
Physiologic Strains. Heating, Piping and Air Conditioning, 1955.
SECTION 7-SAFETY
7-11
-------
The resulting physiological strain is determined by the ratio of
stress (E req.) to the maximum capacity of the environment (E max.)- The
resulting value is defined as the Heat Stress Index (HSI), which is calcu-
lated as:
HSI = c ' * 100
E max.
The values E req. and E max. may be calculated at the maximum exposure
time based on the HSJ defined. Generally, HSI maximum acceptable values
are established for an 8-hour work day.
Table 7-6 indicates expected physiological and hygienic implications
of an 8-hour exposure at various heat-stress levels.
A nomograph may be used to evaluate acceptable exposure times under
various conditions. Figure 7-1 shows the methodology for calculating
exposure time. Constants and variables used in the nomograph are as follows:
R = 17.5 (Tw - 95)
C = 0.756 V°'6(Ta - 95)
E max. = 2.8 V0'6 (42-PWa)
where R = radiant heat exchange, Btu/h
C = convective heat exchange, Btu/h
E max. = max. evaporative heat loss, Btu/h
Tw = mean radiant, temperature, °F
Ta = air temperature, °F
V = air velocity, ft/min
PWa = vapor press., mm Hg
Twb = wet bulb temperature, °F
M = metabolic rate, Btu/h
Tg = globe temperature, °F
SECTION 7-SAFETY
7-12
-------
TABLE 7-6. INDEX OF HEAT STRESS'
Index of
Heat Stress
(HSI)
Physiological and hygienic implications of 8-hour exposures to
various heat stresses
-20
-10
0
+10
20
30
40
50
60
70
80
90
100
Mild cold strain. This condition frequently exists in areas
where persons recover from exposure to heat.
No thermal strain.
Mild to moderate heat strain. Where a job involves higher
intellectual functions, dexterity, or alertness, subtle to
substantial decrements in performance may be expected. When
a job requires heavy physical work, little decrement expected
unless ability of individuals to perform such work under no
thermal stress is marginal.
Severe heat strain, involving a threat to health unless persons
are physically fit. A break-in period is required for those not
previously acclimatized. Some decrement in performance of physi-
cal work is to be expected. Medical selection of personnel is
desirable because these conditions are unsuitable for those
with cardiovascular or respiratory impairment or with chronic
dermatitis. These working conditions are also unsuitable for
activities requiring sustained mental effort.
Very severe heat strain. Only a small percentage of the popu-
lation may be expected to qualify for this work. Personnel
should be selected by medical examination and by trial on the
job (after acclimatization). Special measures are needed to
assure adequate water and salt intake. Amelioration of work-
ing conditions by any feasible means is highly desirable,
and should decrease the health hazard and simultaneously
increase efficiency on the job. Slight "indisposition"
that in most jobs would be insufficient to affect performance
may render workers unfit for this exposure.
The maximum strain tolerated daily by fit, acclimatized, young
persons.
aAdapted from Belding and Hatch, "Index for Evaluating Heat Stress in Terms
of Resulting Physiologic Strains," Heating, Piping and Air Conditioning,
1955.
SECTION 7-SAFETY
7-13
-------
AI» E»ArO"-IW COUVtCTIOII KW KH»T Alf TWERATm
WtOCITT. AVAllAfla (C). T€*ERATU«. TB»«ATUK! MFrtRENC[
'!" <*•«•>. Ituh F (t,;, (t--t,l.
fltur c '
(»> 0* **001 K F
45o[
4O>.
330
300
230-
223
200-
in
160
140
120
100-
90
to
7C
_.
30
45
40
33
30
23
(-1000)-
(-•00)-
1400-1
2000-
1750'
1250
1000;
HO-
600-
^^r
^^*^ 400-
^©r ^^-^50
_^^P -— ^*^^r i?00"
^*****~''^ ^— 123-
^rf*^^^^*** — — _ ~ "^"^" IOQ.
~^^^^^ ^^^^ ^^^\,Jy
*^£^***'^ 10'
^^••* .
SO'
40 :
30'
Broken lines indi- ?0
cote solution to example in text.
. (-600)-
(-500)-
(-400)-
. (-300)-
'. ^'
f
.<-i3or-
(-100)'
(-•0)-
(-50)-
-J-40)<
. (-30)-
(-201-
(-13)-
(-10)-
(-3)-
1300
1000 TO-
KO 15C-
123'
•*00
-500 '*•
•0-
400 ^^'
^ ^^ 60 -
-300 ^^^ 50"
3* .n
^^ 40-
-700 30-
150 -20-
*<^*"^*^'5 u-
TOO ,j 5
-90 10 0;
-50 6 0-
^^^^ .5 n'
»40- ^ ^ 3 „•
4.0-
"30 3.0-
2.5
-20 2.0-
-15 '-»-
-10 '•°°"
0.30
3
. _
-
,
^
«^^
t ^^
^
» ^
50
*0
65
70 (70)-
75
(75)-
•BO
82 (801-
84 ^^*,
186 (15).
88 I")'
(17)-
(Ml-
90 (89).
H— «»•
92 ^'^'
(92)-
93
(93)-
94
140
130 ^
•
120
115
\\&^f
105
104
103
•102
101
100
•9»
98
•97
•96
100
K
60
50
40
30
20
15
to
a o
6 0
5 0
4 0
3 0
2.3
2 0
1.5
1.0
I Broken linei and olio columns are 2 m HI JZ H It
numbered to indicate steps in the
solution.
RA31ATH* HALL NEtAdOLW 6LOEE EVAPORATION AUWABLE HETABOLIS" QHHECT1W EMPOSATIOt,
(R), IEWERATUP.E »K TENttRSTlWE HEOUHEO EXPOSURE (M) 1C). AVAILABLE.
•t"*1 (t.), KADIATIW, ds).t (£rto). T:IE, tw itu* "«,'•
K F *tuh Bluh " N-"£t Ituh
2730-1
•D,
70-
*0'
30'
40
30-
20-
10
0'
2500-
2250-
2125
2000-
1750-
1625
1500-
1375
1250-
1175
1000-
— 875
625
300-
375
250-
125
o-l
L S230i
•—•«.
*~v»
-s.
\
O
1
1
Ov
&
^
^i)
i5
I
x
rZX)
245 500°-
•240 4750 •
J^f 4300-
M3 «*>•
-220 4000-
-210 375°
3300-
3250
•190 3000'
<< 2S°°-
2250
" "£— — — — T —
'50 .730
"14C 1300-
"13° 1230
"l!5 1000-
-100 7*>
" 300-
<
«»^«"
•180 -
175 ',
170 \
163
•160 .
155
,150-*;
145 -
•135 '
1JO
125 •
120
115 •
110
105
100 .
95 J
6750 i
-6300 2-1/4-
•6000 2-1/2-
2-3/4-
-5500
3-
t-5000
3-l/J;
•4500 ^^* 4.
•4000 j.
•3500 f^f£*
-^ "* g.
"^x^ '°:
-2500 **|j.
•2000
120
•1500 1*0
•1000
-300
•0
io-
W;
J
^•*"
^.--
«
3;
X~^
-2300
2400
2X0
2200
2100
•2000
1900
1800
1700^^^
-1500
1400
1300
1200
*"00lS^.
-1000
900 J
800
700
400
-300
•1300
T*"
1300
1200
1100
•1000
900 ^^
KXi^. -^**"
600
•500
400
300
200
100
•0
-100
-200
-300
-400
-SCO
200
400
600
800
1000
1200
1400
1600
1800
•2000
2200
2400
JUi Jf U TIT ^;llf 1111 KIV Tf JJ ^
Figure 7-1. Nomograph developed by McKarns and Brief
incorporating the revised Fort Knox coefficients.
Source: McKarns, J. S., and R. S. Brief. Nomographs Give Refined Estimates
of Heat Stress Index. Heating, Piping and Air Conditioning
38:113, 1966.
7-14
-------
An example presented here illustrates the use of the nomograph under
the following conditions: Tg = 130°F, Ta = 100°F, Twb = 80°F, V = 50
ft/min, M = 2000 Btu/h and dew point = 73°F.
Step 1. Determine convection (c). Connect velocity (column I) with air tem-
perature [(Ta) column II] and read c on column III.
Step 2. Determine E max. Connect velocity (column I) and dew point (column IV)
and read E max. on column V.
Step 3. Determine constant K. Connect velocity (column I) with Tg-Ta (column
VI) and read K on column VII.
Step 4. Determine Tw. Connect K (column VII) and Tg (column VIII) and read
Tw on column IX.
Step 5. Extend line in step 4 to column X and read R.
Step 6. Connect R (column X) with M (column XI) and read R + M on column XII.
Step 7. Connect C (column III) with R + M (column XII) and read E req. on
column XIII.
Step 8. Connect E max. (column V) with E req. (column XIII) and read allowable
exposure time on column XIV.
Metabolic rate varies with exertion and work expended, and an estimate
of M must be made for each effort expended in the fabric filter inspection
or repair. Examples of M for several levels of activity are provided in
Tables 7-7 and 7-8.
TABLE 7-7. HEAT PRODUCTION FOR VARIOUS LEVELS OF EXERTION
Activity
Sleeping
Sitting quietly
Working at a desk, driving a car, standing, minimum
movement
Sentry duty, standing at machine while doing light work
Walking 2.5 mph on level, moderate work
Walking 3.5 mph or level, moderately hard work
Walking 3.5 mph on level with 45 Ib load, hard work
cal/m2h
40
50
80
100
150
200
300
aAdapted from: The Industrial Environment - Its Evaluation and Control.
NIOSH, 1973.
SECTION 7-SAFETY
7-15
-------
TABLE 7-8. BODY HEAT PRODUCTION AS A FUNCTION OF ACTIVITY'
Activity
Rest (seated)
Light machine work
Walking, 3.5 mph on level
Forging
Shoveling
Slag removal
kcal/h
90
200
300
390
450-600
700
Adapted from:
The Industrial Environment - Its Evaluation and Control.
NIOSH, 1973.
When the work involves lifting, pushing, or carrying loads; cranking;
etc., the heat equivalent of the external work (W) is subtracted from the
total energy output to obtain heat produced in the body (M).
SECTION /-SAFETY
7-16
-------
SECTION 8
MODEL O&M PLAN
Generally, one or more individuals at a plant site are responsible for
ensuring that a fabric filter is operated and maintained so that it meets
design removal efficiencies for particulate matter and that the plant com-
plies with regulatory emission limits.
Unfortunately, most O&M personnel do not receive in-depth training on
the theory of fabric filter operation, diagnostic analysis, and the problems
and malfunctions that may occur over the life of a unit. Plant personnel
tend to learn about the operation of a specific unit and to gain operating
experience as a result of day-to-day operating problems. This so-called
"on-the-job" training can result in early equipment deterioration or cata-
strophic failures that could have been avoided.
This section presents the basic elements of an O&M program that will
prevent premature fabric filter failure. This program is not all-inclusive,
and it does not address all potential failure mechanisms. Nevertheless, it
provides the user with enough knowledge to establish a plan of action, to
maintain a reasonable spare parts inventory, and to keep the necessary rec-
ords for analysis and correction of deficiencies in fabric filter operation.
The overall goal of an O&M plan is to prevent unit failures. In case
failures do occur, however, the plan must include adequate procedures to
limit the extent and duration of excess emissions, to limit damage to the
equipment, and to effect changes in the operation of the unit that will
prevent recurrence of the failure. The ideal O&M program includes require-
ments for '-ecordkeeping, diagnostic analysis, trend analysis, process anal-
ysis, and an external and internal inspection program.
The components of an O&M plan are management, personnel, preventive
maintenance, inspection program, specific maintenance procedures, and
SECTION 8-MODEL O&M PLAN
-------
internal plant audits. The most important of these are management and per-
sonnel. Without a properly trained and motivated staff and the full support
of plant management, no O&M program can be effective.
8.1 MANAGEMENT AND STAFF
Personnel operating and servicing the fabric filter must be familiar
with the components of the unit, the theory of operation, limitations of the
device, and proper procedures for repair and preventive maintenance.
For optimum performance, one person (i.e., a coordinator) should be
responsible for fabric filter O&M. All requests for repair and/or investiga-
tion of abnormal operation should go through this individual for coordination
of efforts. When repairs are completed, final reports also should be trans-
mitted to the originating staff through the fabric filter coordinator. Thus,
the coordinator will be aware of all maintenance that has been performed,
chronic or acute operating problems, and any work that is in progress.
The coordinator, in consultation with the operation (process) personnel
and management, also can arrange for and schedule all required maintenance.
He/she can assign priority to repairs and order the necessary repair compo-
nents, which sometimes can be received and checked out prior to installation.
Such coordination does not eliminate the need for specialists (electricians,
pipe fitters, welders, etc.), but it does avoid duplication of effort and
helps to ensure an efficient operation.
Many fabric filter failures and operating problems are caused by mechan-
ical deficiencies. These are indicated by changes in differential pressures
and temperatures and by opacity readings. By evaluating process conditions,
pressure and temperature readings, inspection reports, and the physical
condition of the unit, the coordinator can evaluate the overall condition of
the unit and recommend process modifications and/or repairs.
The number of support staff required for proper operation and mainte-
nance of a unit is a function of unit size, design, and operating history.
Staff requirements must be assessed periodically to ensure that the right
personnel are available for normal levels of maintenance. Additional staff
will generally be needed for such activities as a major rebuilding of the
unit and/or structural changes. This additional staff may include plant
SECTION 8-MODEL O&M PLAN
8-2
-------
personnel, outside hourly laborers, or contracted personnel from service
companies or fabric filter vendors. In all cases, outside personnel should
be supervised by experienced plant personnel. The services of laboratory
personnel and computer analysts may also be needed. The coordinator should
be responsible for final acceptance and approval of all repairs. Figure 8-1
presents the general concept and staff organizational chart for a centrally
coordinated O&M program.
As with any highly technical process, the O&M staff responsible for the
fabric filter must have adequate knowledge to operate and repair the equip-
ment.
Many components of a fabric filter are not unique, and special knowledge
is not required regarding the components themselves; however, the arrangement
and installation of these components are unique in most applications, and
special knowledge and care are necessary to achieve their optimum performance.
Many plants have a high rate of personnel turnover, and new employees
are assigned to work on a fabric filter who may have had no previous contact
with air pollution control equipment. To provide the necessary technical
expertise, management must establish a formal training program for each
employee assigned to fabric filter maintenance and operation.
An optimum training program should include the operators, supervisors,
and maintenance staff. Changes in operation that affect temperature, oil or
moisture content, acid dew point, and the particulate abrasiveness of the gas
stream entering the unit have a detrimental effect on fabric filter opera-
tion. The process operator has control over many of these variables. An
understanding of the cause-and-effect relationship between process conditions
and the fabric filter can help to avoid many performance problems. Safety is
an important aspect of any training program. Each person associated with the
unit should have complete instructions regarding confined-area entry, first
aid, and lock-out/tag-out procedures.
Thus, a typical fabric filter training program should include safety,
theory of operation, a physical description of the unit, a review of subsys-
tems, normal operation (indicators), and abnormal operations (common failure
mechanisms), troubleshooting procedures, a preventive maintenance program,
and recordkeeping.
SECTION 8-MODEL O&M PLAN
8-3
-------
MANAGEMENT
PURCHASING
FABRIC FILTER
COORDINATOR
LABORATORY
MAINTENANCE
SUPERVISOR
PROCESS
SUPERVISOR
ELECTRICAL
FOREMAN
ELECTRICIANS
MECHANICAL
FOREMAN
UNIT
OPERATORS
MECHANICS
Figure 8-1. Organizational chart for centrally coordinated
fabric filter O&M program.
SECTION 8-MODEL O&M PLAN
8-4
-------
The O&M program should emphasize optimum and continuous performance of
the unit. The staff should never get the impression that less-than-optimum
fabric filter performance is acceptable. Redundancy is established in the
unit solely to provide a margin of safety for achieving compliance during
emergency situations. Once a pattern is established that allows a less-than-
optimum condition to exist (i.e., reliance on built-in redundancy), less-
than-optimum performance becomes the norm, and the margin of safety begins to
erode.
To reenforce the training program, followup written material should be
prepared. Each plant should prepare and continually update a fabric filter
operating manual and a fabric filter maintenance manual for each unit. A
generic manual usually is not adequate because each vendor's design philoso-
phy varies. The use of actual photographs, slides, and drawings aids in the
overall understanding of the unit and reduces lost time during repair work.
Training material and courses available from manufacturers and vendors
should be reviewed and presented as appropriate. Further, staff members
responsible for each unit should attend workshops, seminars, and training
courses presented by the Electric Power Research Institute (EPRI), the Port-
land Cement Association (PCA), EPA, and other organizations to increase the
scope of knowledge and to keep current with evolving technology.
8.2 MAINTENANCE MANUALS
Specific maintenance manuals should be developed for each fabric filter
at a source. The basic elements of design and overall operation should be
specific to each fabric filter and should incorporate the manufacturer's
literature and in-house experience with the particular type of unit. The
manual should relate to the physical aspects of the unit. Descriptions
should be brief and to the point; long narratives without direct application
should be avoided.
Figure 8-2 presents a suggested outline for a typical manual. The manu-
al should begin with such basic concepts as fabric filter description and op-
eration. It can then continue with a section on component parts, which
should include detailed drawings and an explanation of the function of each
component and its normal condition.
SECTION 8-MODEL O&M PLAN
8-5
-------
A. FABRIC FILTER DESCRIPTION (GENERAL)
1. Participate Collection System
2. Cleaning System
3. Ash Removal System
B. DESCRIPTION OF OPERATION
1. Collection Mechanisms
2. Filter Cake Removal
C. SAFETY EQUIPMENT
1. Self-Contained Breathing
Apparatus
2. Gas Monitoring Equipment
3. Protective Clothing
4. Eye and Ear Protection
5. Gas Masks with Appropriate
Filters
6. Tags
D. COMPONENT DESCRIPTION
1. Filter Media
2. Cleaning System
3. Housing
4. Valves and Dampers
5. Motors, Fans, and Belts
6. Auxiliary Systems
E. INTERNAL INSPECTION AND MAINTENANCE
1,
Bags
a.
b.
c.
d.
Worn, Abraded, Damaged Bags
Condensation on Bags
Tension
Loose, Damaged, or Improper
Bag Connection
Inlet and Outlet Ducts
a. Dust Buildup
b. Baffle
Hoppers
a. Dust Buildup in Hoppers
b. Hopper Heater Operation
Corrosion on All Surfaces
F. EXTERNAL INSPECTION AND MAINTENANCE
1. Cleaning System
a. Operation Without Binding
b. Loose or Worn Bearings
c. Drive Components
d. Solenoids, Pulsing Valves
(Pulse-Jet)
e. Compressed-Air System
(Pulse-Jet)
f. Damper Valves
2. Air Leakage
a. Expansion Joints
b. Door Gaskets
c. Cleaning System Penetra-
tions
d. Hoppers
3. Interlocks
a. Operation
b. Lubrication
4. Control Cabinet
a. Cleanliness
b. Loose Connections
c. Air Filter
APPENDIX
1.
Inspection and Maintenance
Checklist
2. Layout Details
Figure 8-2. Outline for Fabric Filter Maintenance Manual
SECTION 8-MODEL O&M PLAN
8-6
-------
The next section covers the internal inspection and maintenance procedure,
which is extremely critical in maintaining performance. Periodic checks are
necessary to maintain bag integrity, to remove accumulated ash deposits, and to
prevent air inleakage. The section on external inspection and maintenance
includes all supporting equipment, such as cleaning mechanisms, instrumentation,
air compressors (where applicable), etc. Each of these sections should provide a
procedure for evaluating the component. The manual should identify key operating
parameters, define normal operation, and identify indicators of possible
deviations from normal condition. Key operating parameters include temperature,
pressure, cleaning cycle, opacity, or other parameters that can be used to estab-
lish the basic operating condition of the unit.
After evaluation of conditions, a procedure must be presented to replace,
repair, or isolate each component. Unless a proper procedure is followed,
the corrective action could result in further damage to the unit, excessive
emissions, or repeated failure.
8.3 OPERATING MANUALS
Whereas maintenance manuals are designed to facilitate physical repairs
to the fabric filter, operating manuals are needed to establish an operating
norm or baseline for each unit. Maintenance of the physical structure cannot
ensure adequate performance of the unit because gas stream conditions such as
temperature, gas composition, and gas volume can cause premature bag failure
and rapidly decrease collection efficiency.
The operating manual should parallel the maintenance manual in terms of
introductory material so that the operators and maintenance personnel have
the same basic understanding of the components and their function and, of the
overall operating theory. Additional information should be provided on the
effects of major operating variables such as gas volume, gas temperature, and
pressure drop. The manual also should discuss the effects of air inleakage on
the bags, potential condensation problems, and the points where inleakage may
occur (hoppers, doors, expansion points, etc.). Figure 8-3 presents an
outline for an operating manual.
SECTION 8-MODEL O4M PLAN g-7
-------
DESCRIPTION OF FABRIC FILTERS
1. Particulate-Collection System
2. Cleaning System
3. Dust Removal System
DESCRIPTION OF OPERATION
1. Collection Mechanisms
2. Filter Cake Removal
OPERATIONAL FACTORS
1. Gas Volume
a. Excess Air
b. Air Inleakage
(1) Hoppers
(2) Access Doors
(3) Expansion Joints
(4) Test Ports
(5) Process Points
2. Gas Temperature
a. High Temperature
b. Low Temperature
(1) Acid and Moisture
Dewpoint
3. Differential Pressure
a. High
b. Low
4. Opacity
ASH-REMOVAL-SYSTEM MALFUNCTION
1. Plugged Hopper
2. Low Vacuum
a. Excess Air Inleakage
b. Valves Stuck Open
STARTUP
1. Safety Check
2. Cleaning System On
3. Ash Removal System On
4. Hopper Heaters On
SHUTDOWN
1. System Purged
2. Hoppers Emptied
3. Cleaning System Turned Off
4. If Long Outage, Compressor,
Hopper Heaters, and Dust-
Removal System Turned Off
Figure 8-3. Fabric Filter Operating Manual outline.
(Courtesy of PEI Associates, Inc.)
SECTION 8-MODEL O&M PLAN
8-8
-------
With regard to fuel combustion sources, the manual should discuss the
effects of such process variables as burner conditions, burner alignment, and
pulverizer fineness, which change the ash particle properties and size
distribution. An expected normal range of values and indicator points should
be established as reference points for the operator.
Startup and shutdown procedures should be established, and step-by-step
instructions should be provided to ensure sequenced outage of equipment to
aid in maintenance activities and to eliminate startup problems.
8.4 SPARE PARTS
An inventory of spare parts should be maintained to replace failed parts
as needed. Because all components or subassemblies cannot be stocked, a
rational system must be developed that establishes a reasonable inventory of
spare parts. Decisions regarding which components to include in the spare
parts inventory should be based on the following:
1. Probability of failure
2. Cost of components
3. Replacement time (installation)
4. Whether the part can be stored as a component or subassembly (i.e.,
shaker assembly vs. individual components)
5. In-house technical repair capabilities
6. Available space
The probability of failure can be developed from outside studies (e.g.,
EPRI), vendor recommendations, and a history of the unit. It is reasonable
to assume that components subjected to heat, dust, weather, or wear are the
most likely to fail. Components of this type are no different from those in
process service, and reasonable judgment must be used in deciding what to
stock. Maintenance staff members should be consulted for recommendations
concerning some items that should be stocked and the number required. Ad-
justments can be made as operating experience is gained. Items that fall
into this category include solenoids, drive belts, tension springs, shaker
motor drives, and level indicators.
SECTION 8-MODEL O&M PLAN o
-------
Another factor in defining a spare parts inventory is the cost of indi-
vidual components. Although stocking bags or door seal components are not
costly, stocking a spare compressor can be quite costly. Maintaining an
extensive inventory of high-cost items that have low probability of failure
is not justified.
The time required to receive the part from the vendor and the time re-
quired to replace the part on the unit also influence whether an item should
be stocked. If the lead time for a critical part is a matter of weeks or
months, or if a component must be specially built, stocking such items is
advantageous.
Many plants have an electronics and mechanical shop whose highly trained
staff can repair or rebuild components to meet original design specifica-
tions. The availability of this service can greatly reduce the need to
maintain component parts or subassemblies. In these cases, one replacement
can be stocked for installation during the period when repairs are being
made. For example, many printed circuit boards can be repaired internally,
which reduces the need to stock a complete line of electronic spare parts.
8.5 WORK ORDER SYSTEMS
A work order system is a valuable tool that allows the fabric filter
coordinator to track unit performance over a period of time. Work order and
computer tracking systems are generally designed to ensure that the work has
been completed and that charges for labor and parts are correctly assigned
for accounting and planning purposes. With minor changes in the work order
form and in the computer programs, the work order also can permit continuous
updating of failure-frequency records and can indicate whether the mainte-
nance performed has been effective in preventing repeated failures. In
general, the work order serves three basic functions:
1. It authorizes and defines the work to be performed.
2. It verifies that maintenance has been performed.
3. It permits the direct impact of cost and parts data to be entered
into a central computerized data handling system.
SECTION 8-MODEL O&M PLAN
8-10
-------
To perform these functions effectively, the work order form must be specific,
and the data fields must be large enough to handle detailed requests and to
provide specific responses. In many computerized systems, the data entry
cannot accommodate a narrative request and specific details are lost.
Most systems can accommodate simple repair jobs because they do not
involve multiple repairs, staff requirements, or parts delays. Major re-
pairs, however, become lost in the system as major events because they are
subdivided into smaller jobs that the system can handle. Because of this
constraint, a large repair project with many components (e.g., a cleaning
system failure or control panel repair) that may have a common cause appears
to be a number of unrelated events in the tracking system.
For diagnostic purposes, a subroutine in the work order system is nec-
essary that links repairs, parts, and location of failure in an event-time
profile. Further, the exact location of component failures must be clearly
defined. In effect, it is more important to know the pattern of failure than
the cost of the failure.
The goel of the work order system can be summarized in the following
i terns:
0 To provide systematic screening and authorization of requested
work.
0 To provide the necessary information for planning and coordination
of future work.
0 To provide cost information for future planning.
0 To instruct management and craftsmen in the performance of repair
work.
0 To estimate manpower, time, and materials for completing the re-
pair.
0 To define the equipment that may need to be replaced, repaired, or
redesigned (work order request for analysis of performance of
components, special study, or consultation, etc.).
Repairs to the unit may be superficial or cosmetic in nature or they may
be of an urgent nature and require emergency response to prevent damage or
failure. In a major facility, numerous work order requests may be submitted
as a result of daily inspections or operator analysis. Completing the jobs
SECTION 8-MODEL O&M PLAN
8-11
-------
in a reasonable time requires scheduling the staff and ordering and receiving
parts in an organized manner.
For effective implementation of the work order system, the request must
be assigned a level of priority as to completion time. These priority as-
signments must take into consideration plant and personnel safety, the poten-
tial effect on emissions, potential damage to the equipment, maintenance per-
sonnel availability, parts availability, and boiler or process availability.
Obviously, all jobs cannot be assigned the highest priority. Careful as-
signment of priority is the most critical part of the work order system, and
the assignment must be made as quickly as possible after requests are re-
ceived. An example of a five-level priority system is provided in Figure
8-4.
If a work order request is too detailed, it will require extensive time
to complete. Also, a very complex form leads to superficial entries and
erroneous data. The form should concentrate on the key elements required to
document the need for repair, the response to the need (e.g., repairs com-
pleted), parts used, and manpower expended. Although a multipage form is
not recommended, such a form may be used for certain purposes. For example,
the first page can be a narrative describing the nature of the problem or
repair required and the response to the need. It is very important that the
maintenance staff indicate the cause of the failure and possible changes that
would prevent recurrence. It is not adequate simply to make a repair to
malfunctioning shaker cleaning system controls and respond that "the repairs
have been made." Unless a detailed analysis is made of the reason for the
failure, the event may be repeated several times. Treating the symptom
(making the repair; replacing bags, solenoid valves, etc.) is not sufficient;
the cause of the failure must be treated.
In summary, the following is a list of how the key areas of a work order
request are addressed:
1. Date - The date is the day the problem was identified or the job
was assigned if it originated in the planning, environmental, or
engineering sections.
2. Approved by - This indicates who authorized the work to be com-
pleted, that the request has been entered into the system, and that
it has been assigned a priority and schedule for response. The
SECTION 8-MODEL O&M PLAN
8-12
-------
WORKORDER PRIORITY SYSTEM
PRIORITY ACTION
1 Emergency Repair
2 Urgent repair to be completed
during the day
3,4 Work which may be delayed
and completed in the future
5 Work which may be delayed
until a scheduled outage
Figure 8-4. Example of five-level priority system.
(Courtesy of PEI Associates, Inc.)
8-13
-------
3.
4.
5.
6.
maintenance supervisor or fabric filter coordinator may approve the
request, depending on the staff and the size of the facility. When
emergency repairs are required, the work order may be completed
after the fact, and approval is not required.
Priority - Priority is assigned according to job urgency on a scale
of 1 to 5.
Work order number - The work order request number is the tracking
control number necessary to retrieve the information from the
computer data system.
Continuing or related work order numbers - If the job request is a
continuation of previous requests or represents a continuing prob-
lem area, the related number should be entered.
Equipment number - All major equipment in a fabric filter should be
assigned an identifying number that associates the repair with the
equipment. The numbering system can include process area, major
process component, fabric filter number, fabric filter compartment,
equipment number, and component. This numeric identification can
be established by using a field of grouped numbers. For example,
the following could be used:
ID number
XX - XXX - XX - XXX - XXX - XXX
- component
equipment number
fabric filter compartment
L-fabric filter number
process subcomponent
1—process area
If the facility only has one fabric filter and one process, the
first five numbers (two groups) may not be required, and the entry
is thus simplified. The purpose of the ID system is to enable
analysis of the number of events and cost of repair in preselected
areas of the fabric filter. The fineness or detail of the equip-
ment ID definition will specify the detail available in later
analyses.
7. Description of work - The request for repair is usually a narrative
describing the nature of the failure, the part to be replaced, or
the work to be completed. The description must be detailed but
SECTION 8-MODEL O&M PLAN
8-14
-------
brief because the number of characters that can be entered into the
computerized data system is limited. Additional pages of lengthy
instruction regarding procedures may be attached to the request
(not for computer entry).
8. Estimated labor - Assignment of personnel and scheduling of outages
of certain equipment require the inclusion of an estimate of man-
hours, the number of in-house staff needed, and whether outside
labor is needed. The more complex jobs may be broken down into
steps, with different personnel and crafts assigned specific re-
sponsibilities. Manpower and procedures in the request should be
consistent with procedures and policies established in the O&M
manual.
9. Material requirements - In many jobs, maintenance crews will remove
components before a detailed analysis of the needed materials can
be completed; this can extend an outage while components or parts
are ordered and received from vendors or retrieved from the spare
parts inventory. Generally, the cause of the failure should be
identified at the time the work order request is filled, and spe-
cific materials needs should be identified before any removal ef-
fort begins. If the job supervisor knows in advance what materials
are to be replaced, expended, or removed, efficiency is increased
and outage time reduced. Also, if parts are not available, orders
may be placed and the parts received prior to the outage. Material
requirements are not limited to parts; they also include tools,
safety equipment, etc.
10. Action taken - This section of the request is the most important
part of the computerized tracking system. A narrative description
of the repair conducted should be provided in response to the work
order request. The data must be accurate and clearly respond to
the work order request.
11. Materials replaced - An itemized list of components replaced should
be provided for tracking purposes. If the component has a pre-
selected ID number (spare parts inventory number), this number
should be included.
Actual man-hours expended in the repair can be indicated by work order
number on separate time cards and/or job control cards by craft and personnel
number.
Copies of work orders for the fabric filter should be retained for
future reference. The fabric filter coordinator should review these work
orders routinely and make design changes or equipment changes as required to
reduce failure or downtime. An equipment log also should be maintained, and
SECTION 6-MODEL O4M PLAN
8-15
-------
the work should be summarized and dated to provide a history of maintenance
on the unit.
Figure 8-5 shows a simplified work order request form. Changes in
design for individual applications and equipment must be made to meet site-
specific requirements.
8.6 COMPUTERIZED TRACKING
8.6.1 Work Orders
If the work completed and parts used in the fabric filter have been
entered in the computerized work order system with sufficient detail, mainte-
nance and management personnel can evaluate the effectiveness of fabric
filter maintenance.
Preventive maintenance (PM) man-hours versus repair man-hours also can
be compared to evaluate the effectiveness of the current PM program. The
level of detail may allow tracking of the impact of PM on particular
subgroups (e.g., shakers, hoppers) as changes are made in PM procedures. The
effectiveness of the PM program may be further evaluated by the required
number of emergency repairs versus scheduled repairs over a period of time
(i.e., priority 2 versus priority 5, etc.).
It should be emphasized that the purpose of the computerized tracking
system is not to satisfy the needs of the accountants or programmers or to
state that the plant has such a system. Rather, the purpose of a computer-
ized tracking system is to provide the necessary information to analyze
fabric filter maintenance practices and to reduce component failures and
excess emissions. The maintenance staff and fabric filter coordinator must
clearly define the kind of data required, the level of detail, and the type
of analysis required prior to the preparation of the data-handling and
report-writing software. Examples of output may be man-hours by department,
man-hours by equipment ID, number of repairs, number of events, number of
parts, and frequency of events.
8.6.2 Fabric Filter Operating Parameters
In addition to tracking work orders, the computer can be used to develop
correlations between process and fabric filter parameters and observed
SECTION 8-MODEL O&M PLAN
8-16
-------
WORK ORDER
FOSSIL STATION
WORK REQUEST
ORIGINATOR
T UNIT | PRIORITY
I I ! I !
LOCATION - EQUIPMENT
AVAILABLE • DATE
REQ'O COM?
DATE
] RED TAG
APPROVED 8V
DATE
I I I I
EQUIP NO
_l ..I i
EQUIPMENT NAME OR JOB TITLE
i i i i—I—L—I—I—I—I—I—I—i ' i
CHARGE TO:
DESCRIPTION OF JOB
i 1 1 I I i I I 1 I I j
CLASS
_J I
ESTIMATED LABOR
DESCRIPTION OF WORK BY CRAFT SKILLS
SEQ.
CRAFT
SKILL
MEN X HOURS
TOTAL
ESTIMATED
MAN/HOURS
SAFETY PROCEDURES:
SAFETY EQUIPMENT REQUIRED:
MATERIAL REQUIREMENT
DELIVER TO
DESCRIPTION
OTY AVAIL USED
SPECIAL EQUIP REQUIRED
ACCEPTED BY
MAINTENANCE SUPERVISOR
CODE
-Figure 8-5a. Example of work order form.
(Copyright®April 1983, EPRI Report CS-2908, "Proceedings: Conference on
Electrostatic Precipitation Technology for Coal Fired Plants". Reprinted
with permission. ) 0 ,.,
o-1 /
-------
UNIT
SYSTEM
SUBSYSTEM
COMPONENT
SUBCOMPONENT
MAINTENANCE REQUEST "CRN
000000
ORIGINATOR:
ASSIGNED TO:
PRIORITY:
DATE;
TINE:
UNIT STATUS:
PROBLEM DESCRIPTION:
FOREMAN:
DATE:
JOB STATUS:
CAUSE OF PROBLEM:
DONE:
SUPERVISOR-,
COMPLETION HATE:.
MATERIALS USED:
TOTAL ^ANHOURS
.IATESIAL COST
.-.Figure 8-5b. Example of work order form.
(Copyright^ April 1983, EPRI_Report CS-2908, "Proceedings: Conference on
Electrostatic Precipitation technology for Coal Fired Plants". Reprinted
with permission.) ' 8-18
-------
emission profiles. Depending on the type of cycles expected in process
operation, the data may be continuously input into the system or it may be
entered from operating logs or daily inspection reports once or twice a week.
The key data for tracking performance are pressure differentials, opaci-
ty (i.e., 6-minute averages), boiler load (or associated parameter propor-
tional to gas flow volume), flue gas temperature, and fuel quality data
(i.e., fuel source, ash, fineness, etc.).
8.7 PROCEDURES FOR HANDLING MALFUNCTION
Many malfunctions are of an emergency nature and require prompt action
by maintenance staff to reduce emissions or prevent damage to the unit. On
some units, predictable but unpreventable malfunctions can be identified;
such malfunctions include hopper pluggages, bag failure, and cleaning system
failure. These problems, as well as corrective actions, are discussed in
Sections 4.2 and 4.3.
An effective O&M program should include established written procedures
to be followed when malfunctions occur. Having a predetermined plan of
action reduces lost time, increases efficiency, and reduces excessive emis-
sions. The procedures should contain the following basic elements: malfunc-
tion anticipated, effect of malfunction on emissions, effect of malfunction
on equipment if allowed to continue, required operation-related action, and
maintenance requirements or procedure.
SECTION 8-MODEL O&M PLAN
8-19
-------
REFERENCES FOR SECTION 8
1. Vuchetich, M. A., and R. J. Savoi. Electrostatic Precipitator Training
Program and Operation and Maintenance Manual Development at Consumers
Power Company. In: Proceedings Conference on Electrostatic Precipi-
tator Technology For Coal-Fired Power Plants. EPRI CS-2908 - April
1983.
2. Rose, W. 0. Fossil Maintenance Documentation at Duke Power Company.
In: Proceedings Conference on Electrostatic Precipitator Technology For
Coal-Fired Power Plants. EPRI CS-29C8 - April 1983.
8-20
SECTION 8-MODEL O&M PLAN
-------
APPENDIX A
EXAMPLES OF FABRIC FILTER O&M FORMS
A-l
APPENDIX A - EXAMPLES OF FABRIC FILTER O&M FORMS
-------
m
g
x
m
x
T3
m
OD
2
o
T1
f-
-I
m
O
2
O
3)
2
Company
Type of
inspection
Check
Good
Needs
attn. Item
Serial No.
( ) Structural -bolts ( )
( ) Ladder assembly ( )
( ) Airlock ( )
( ) Drive assembly
A. Gear reducer ( )
B. Drive shaft align ( )
C. Coupler shaft ( )
D. Bearings ( )
E. Belts ( )
F. Sheaves ( )
G. Serial No. (motor)
H. hp
I. rpm
J. Sheave size
( ) Transfer screw assy. ( )
( ) Fan
A. Serial No.
B. Model No.
C.
D. Make
E. Sheaves ( )
F. Sheave diameter ( )
G. Shaft diameter ( )
H. Series of belts
15. ( )
( ) 16. ()
( ) 17. ()
( ) 18. ()
( ) 19. ()
( ) 20. ()
( ) 21. ()
22. ( )
23. ( )
24. ( )
( ) 25. ()
I . Shaft diameter
J. Make
( ) Water trap ( )
( ) Air regulator ( )
( ) Bin indicator ( )
( ) Magnehelic ( )
( ) Magnehelic tubing ( )
( ) Baffle wear ( )
( ) 26. ( )
( ) 27. ()
( ) 28. ()
{ ) 29. ()
( ) 30. ()
Check
Good
Top door hold-down
straps
Top door leaks
Manifold pipes
anchored
Manifold pipes holes
Manifold pipes center
over venturi
Venturi properly
seated
Cages properly
installed
Bag clamps R
Bags-Visolite
} Bags-general appear.
Service module
A. Wire connections
terminal box
B. Wire connections
C. Diaphragm valves-
leaks
D. Solenoid valves
operating
E. Hoses and clamps
F. Air pressure leak
psi
Pulse control panel
) Control panel
Dialatrol
Thermocouple
Thermocouple wiring
( )
( )
( )
( )
Date
Name
Needs
attn. Item
( ) ( )
( ) ( )
( ) (
( ) "
( )
( )
( )
31. (
32.
Check Good
) Timer settings
A. Duration
B. Interval
) Delta P
A. Low
High"
Needs
attn.
33.
34.
35.
36.
B. _
C. Average
psi at header
A. Low
B. High"
Average temperature in
baghouse
Visual stack emission
Clean
Dirty
( ) Paint
A. Interior ( ) ( )
B. Exterior ( ) ( )
NOTES
Source:
Example fabric filter inspection report form.
Reigel, S. A. Fabric Filtration Systems, Design, Operation, and Maintenance.
-------
Source I.D. No. SIC _
Inspectors) Date
Inspection Announced?
A. GENERAL PLANT DATA FROM AGENCY FILE
1. Source name, address, and phone number
2. Type of process
3. Allowable emission rate and opacity
4. Date baghouse installation approved
5. Prior complaints or episodes of excess emissions
6. Last inspection date
7. Purpose of inspection
B. GENERAL OBSERVATIONS PRIOR TO ACTUAL INSPECTION
1. Weather conditions
2. Visible emissions
Fabric filter inspection report form.
A-3
-------
3. Is inspector a certified smoke reader? Yes No If yes, give
certification date
(Attach copy of Method 9, if performed)
C. PROCESS INFORMATION
1. Confidential? Yes No
2. Person contacted at plant and title
3. Product(s) produced
4. Production rate(s)
5. Raw materials used
6. Portion of process controlled by baghouse
7. Average uncontrolled emission rate or concentration (indicate
weather obtained from stack test, mass balance, AP-42 emission fac-
tor, other, etc.)
8. Date of last stack test and average emission rate obtained
9. Is cleaned effluent recirculated back into plant? Yes No
D. DUST CHARACTERISTICS (PRIOR TO CONTROL)
1. Is material toxic or otherwise hazardous or does it require special
handling: Yes No Describe
Fabric filter inspection report form. (continued)
A-4
-------
2. Moisture content or other gaseous constituents
3. Abrasiveness or other properties
4. Particle size data - indicate how measured
COLLECTION SYSTEM(S)
1. Baghouse No. 1 No. 2 No. 3
a.
b.
c.
d.
e.
f.
g-
2. Fan
a.
b.
c.
d.
e.
f.
Manufacturer
Type or trade name
Model No.
No. of compartments
Bags/compartment
Bag 1 x d
Total Cloth Area
No. 1 No. 2 No. 3
Manufacturer
Model No.
Blade type
Belt or direct drive
Power rating
Positive or negative
pressure
3. Fabric No. 1 No. 2 No. 3
a.
b.
c.
d.
e.
f.
Manufacturer
Material
Woven or felted
Weave
Weight
Permeability
Fabric filter inspection report form. (continued)
A-5
-------
No. 1 No. 2 No. 3
g. Operating temp, range
h. Surface treatment
i. Coating upon startup
j. Guaranteed life
k. Actual life
4. Cleaning System No. 1 No. 2 No. 3
a. Method
b. Frequency
c. Actuated by
d. Anticollapse rings
e. Wire mesh cages
F. DUST HANDLING SYSTEM(S)
1. Do baghouse hoppers have:
a. Heaters
b. Insulation
c. Level indicators
d. Vibrators
2. Type of dust transport system
3. Fate of collected material
G. INSTRUMENTATION
Do system monitors record any of the following:
1. Process start-up/shutdown
Fabric filter inspection report form. (continued)
A-6
-------
2. System flow or velocity
3. Fan motor amps
4. Temperature (recording?)
5. Pressure
6. Opacity
7. Outlet emissions
8. Compartments off-line
9. Compartments being cleaned
10. Compartments in operation
11. Other
H. OPERATING PARAMETERS - DESIGN AND ACTUAL
Design Actual
1. Flow rate
2. Pressure drop,
flange-to-flange
measurement location
3. A/C, gross
4. A/C, net
(2 comp. down)
5. Temperature
6. Efficiency
Fabric filter inspection report form. (continued)
A-7
-------
7. Emission rate
8. Opacity
I. OPERATING EXPERIENCE/MAINTENANCE ASPECTS
1. Percent of time baghouse fully operational when process is in opera-
tion
2. Has a detailed maintenance schedule been instituted?
3. Is maintenance scheduled as recommended by baghouse manufacturer or
by plant?
4. Are maintenance records available for inspection?
5. How long are records kept on file?
6. Which of the following problem areas have led to periods of excess
emissions or caused the process to be shut down?
Problem Area Duration Frequency
a. Insufficient dust pickup and/or
transport (fugitive emissions)
b. Duct abrasion or corrosion
c. Temperature excursions, high
or low
d. Moisture
e. Fan abrasion, vibration, etc.
f. Gross bag failure
g. Inadequate bag tension
h. Bag chafing or abrasion
Fabric filter inspection report form. (continued)
A-8
-------
Problem Area Duration Frequency
i. Pressure loss
j. Compartment isolation
dampers
k. Cleaning mechanism
1. Visible emissions
m. Plugged hoppers
n. Hopper fires
o. Dust discharge system
CONCLUSIONS/RECOMMENDATIONS
1. Compliance status
2. Need for further action
3. Corrective actions to be taken
4. Time required to rectify problems
5. Special waivers or review of compliance criteria required
6. Need for follow-up inspection
7. Inspector's signature
date
approved by
title
Fabric filter inspection report form. (continued)
A-9
-------
K. OTHER NOTES, COMMENTS, SKETCHES (ATTACH ADDITIONAL PAGES, IF NECESSARY)
Schematic drawings showing locations of process and dust control equip-
ment should be prepared, particularly so, where verbal descriptions may
lead to misunderstandings.
Fabric filter inspection report form.
A-10
-------
BAG FAILURE LOCATION RECORD
ABC
.000
.000
,000
»ooo
»ooo
,000
.000
,000
.000
,000
• ooo
,000
,000
.000
DBF
OOO
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
ooo
G H I
12
11
10
ACCESS DOOR
MODULE NO.
OOO
ooo
ooo
ooo
ooo.
ooo.
OOO-
ooo-
ooo>
ooo-
ooo.
ooo.
ooo
DATE
REPLACE - R
PATCH - P
CAP-OFF - C
RETENSIOH - T
A-11
-------
INTRODUCTION
APPENDIX B
OPERATION AND MAINTENANCE
OF UTILITY FABRIC FILTERS
Interest in the application of baghouse technology to electric utility
boilers began in the late 1960's. The advent of the Clean Air Act of 1970
gave impetus to the investigation of this technology, which continued through
the 1970's. The Clean Air Act also precipitated the particulate emissions
limitations of the 1971 New Source Performance Standards (NSPS), which were
revised in 1979 to include even more stringent particulate emission limita-
tions for utility coal-fired boilers.
Prior to the 1970's, utilities primarily used electrostatic precipitators
(ESP's) for particulate control. These devices were relatively economical
and performed well in terms of particulate removal efficiency with the high-
sulfur (2 to 5 percent) Midwestern and Easter coals common at the time. The
low-sulfur western coals that were used in the western part of the country,
however, produced an ash that was more difficult for an ESP to collect. As a
result, ESP's were less attractive for these applications both in terms of
cost and removal performance. Also, when the more stringent regulatory
standards of the 1970's put strict limitations on S02 emissions, the industry
began shifting from high-sulfur to low-sulfur coals (less than 1 percent) to
reduce S02 emissions. Although the 1971 NSPS regulations for SO^ emissions
could be met with low-sulfur "compliance coal," various utilities began to
use S0? scrubbers when the use of low-sulfur coal was impractical or where,
for example, state S0? emission standards were more stringent than the appli-
cable Federal standards (NSPS).
While these events were taking place, fabric filter technology continued
to evolve. The 1979 NSPS revisions eliminated the advantage of using low-
sulfur coal (at least in the Eastern and Midwestern parts of the country)
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-l
-------
when it became clear that all coal-fired units would be required to employ
some type of flue gas desulfurization (FGD) system regardless of the coal
sulfur content. In addition, these new standards further reduced allowable
particulate emissions, which made ESP's even less practical for low-sulfur
coal applications. About this time, a special class of FGD system called a
spray dryer became available for use with low-sulfur coal. In these systems,
fabric filters are normally used in conjunction with the spray dryer equip-
ment. By combining SO- and particulate collection, this system configuration
reduced the need for separate SOp removal equipment (which can account for as
much as 25 percent of the total plant cost). The first full-scale low-sulfur
application of this type was the 440-MW Coyote power station owned and oper-
ated by Otter Tail Power, which began operations in April 1980.
The traditional barrier to the use of fabric filters in the electric
utility industry was the unavailability of a bag fabric durable enough to
withstand elevated operating temperatures; to resist chemical attack; and to
maintain dimensional stability, tensile strength, and flex strength. When
suitably finished, woven fiberglass fabrics became available in the early
1960's, the use of fabric filters in the utility industry became more feasible.
In 1961, Pennsylvania Power & Light Co. began operations at a pilot
2
fabric filter installation. Although the results were good, the utility
opted for ESP's at the full-scale facility. In 1964, Public Service Electric
& Gas Co. also tested and discarded the fabric filter concept.
Installation of the first full-scale utility fabric filter in the United
States was in 1965—at the 320-MW oil/gas-fired Alamitos station owned and
operated by Southern California Edison Co. A fabric filter was installed at
this facility to eliminate both a visible plume attributed to fine particles
and a sulfuric acid fume resulting from the combustion of residual fuel oil
with a 1.7 percent sulfur content. In this installation, various alkaline
additives (e.g., dolomite and limestone) were injected upstream of the fabric
filter to react with the S03 in the flue gas. This material was then col-
lected in the fabric filter system. After 5 years of operation, the system
was shut down permanently when the utility was unable to obtain a variance to
continue burning high-sulfur oil.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-2
-------
The first full-scale application of a fabric filter at a coal-fired
utility boiler occurred in February 1973. The site of this installation was
the four-boiler, 87.5-MW Sunbury station of Pennsylvania Power & Light Co.
In the 10 years following this initial application, utility commitments to
baghouse technology grew rapidly. By the first quarter of 1984, more than
110 baghouses were either in operation, under construction, or in the design
phase; the total power generating capacity involved was more than 20,000 MW.
The growth in fabric filter usage is illustrated in Figure B-l, which
plots the cumulative installations (in terms of associated electrical gener-
ating capacities) by year of startup. The actual units represented by these
capacity figures are shown in Table B-l. To put this information in proper
perspective, Figure B-2 presents a rescaled version of the Figure B-l plot
superimposed on a plot of the U.S. utility coal-fired power generating capa-
city installed by year.
When compared with the population of coal-fired units as a whole, the
impact of fabric filters is small; however, the number of projected coal-
fired boilers that will be equipped with fabric filters is expected to in-
crease. Also, a significant number of existing plants are expected to convert
to the use of fabric filter technology for particulate emission control in
the years to come.
TYPES OF FABRIC FILTERS IN USE
Fabric filters are normally classified by fabric cleaning method. The
three primary cleaning methods are shake-deflate, reverse-gas, and pulse-jet.
Only the first two are widely used in utility applications, and the reverse-
gas method is by far the most prevalent*. Of the 72 utility boilers equipped
with fabric filters as of June 1981, 9 were of the shake-deflate design, 2
were of the pulse-jet design, and ell the rest were of the reverse-gas design.
Pulse-jet fabric filters work well for the shorter, small-diameter bags
found on smaller-scale industrial applications. Because utility systems
*Recent research and economic studies now show that a shake/deflate fabric
filter with an air-to-cloth ratio of 2.7 acfm/ft2 offers a lower total cost
than reverse-gas units with comparable or lower air-to-cloth ratios.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-3
-------
1972 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
YEAR OF STARTUP
Figure B-l. Cumulative electrical generating capacity controlled by
fabric filters, by year of startup.
B-4
-------
TABLE B-1. FABRIC FILTERS IN OPERATION, UNDER CONSTRUCTION, OR IN THE
DESIGN PHASE IN THE U.S. ELECTRIC UTILITY INDUSTRY
UTILITY
STATION NAME MTINC
NV
MILER
TrPE '
A/C
RATIO
CLEANING
HE I HOC «
DATE
COMMISSIONED
(•cfm/ft*)11
Aruona Public Service
Atlanta City Electric
Atlanta City Electric
Baltimore Gas I Electric
Basin Electric Power Coop.
Cajun Electric Coop,
City of Colorado Springs
City of Colorado Springs
City of Columbia Water and Light
City of Columbia Mater and Light
City of Duluth
Colorado - Ute Electric Assn., Inc.
Colorado - Ute Electric Assn., Inc.
Colorado - Ute Electric Assn., Inc.
Cooperative Power Association
Cooperative Power Association
Crisp County Power Coflimsslon
Dayton Power £ Lignt Company
Fremont Dept. of Utilities
Fremont Dept. of Utilities
Golden Valley Electric Assn., Inc.
Houston Lighting a Power Company
Independence Power & Light Dept.
Kansas City Board of Public Utilities
Kansas City Board of Public Utilities
Marquette Board of Light and Power
Marquette Board of Light and Power
Marquette Board of Light and Power
Marshall Municipal utilities
Marshall Municipal utilities
Minnesota Power and Light Company
Nebraska Public Power District
Nebraska Public Power District
Nevada Power Company
Nevada Power Company
Northern States Power Company
Ohio Edison
Otter Tall Power Company
Pennsylvania Power & Light Company
Pennsylvania Power & Light Company
Pennsylvania Power & Light Company
Pennsylvania Power a Light Company
Philadephia Electric Company
Piqua Mumciple Power System
Plains Electric Gen. Transmission Coop.
Platte River Power Authority
Public Service Co. of Colorado
Public Service Co. of Colorado
Public Service Co. of Colorado
Public Service Co. of Colorado
Public Service Co. of Colorado
Rochester Public Utility Dept.
Sierra Pacific Power Company
Sierra Pacific Power Company
Southern Colorado Power Oiv.
Southwestern Public Service Company
Southwestern Public Service Company
Southwestern Public Service Company
Southwestern Public Service Company
Sunflower Electric Coop.
Tennessee Valley Authority
Texas Utilities Company
Tuscon Electric Company
Tucson Electric Power
Tucson Electric Power
Tucson Electric Power
United Power Association
United Power Association
United Power Association
Utah Power 1 Light
Four Corners 2*800
Deepwater 2x23
Oeepwater 40
Crane 2x197
Antelope Valley 2x440
Oxbow 540
R. D. Ninon 200
Martin Drake 85
Columbia 16. S
Columbia 22
Ouluth 3x--
Bullock 2x6. 25
Kucla 3x13
Craig 440
Coal Creek 91
Coal Creek 92
Plant Crisp 12.5
Longwortn Steam
Lon 0. Wright 22
Lon D. Wright 16. S
Healy ?.c
U. A. Parish 551
Missouri City 2«22
Intermountaln 4x820
Kaw 2>44
Kaw 68
Shiras 15
Shiras n
Shiras 44
Marshall 6
Marshall 16.5
Clay Boswell 2x69
Kramer 3x23
Kramer 36
Reid Gardner 250
Warner Valley 2x250
Riverside 2x110
V. M. Samnis 4x180
Coyote 410
Sunbury 2x87.5
Holtwood 79
Holtwood 79
Brunner Island 350
Cromby 15
Piqua 3x44
Escalante 210
•Rawhide 250
Arapahoe 44
Cameo 22
Cameo 44
Cherokee 110
Cherokee 150
Rochester 2x--
North Valmy 250
North Valmy 250
Clark 16.5
Clark 22
Ray Tolk 2x500
Harrington 350
Harrington 350
Celanese (Cogen.) 2x30
HoUomb 280
Shawnee 10>17S
Montkello 2x575
Springerville 2x350
Irvington 2x50
Irvlngton 90
irvmgton 100
Elk River 2x11.5
Elk River 23
Stanton 172
Hunter 2x400
PC
PC
PC
C
PC
PC
PC
PC
S
S
PC
PC
S
PC
So
So
PC
S
PC
PC
PC
PC
PC
PC
pf
r\,
PC
PC
S
S
PC
PC
PC
PC
PC
PC
PC
PC
PC
PC
C
PC
PC
PC
PC
PC
S
PC
PC
PC
PC
PC
PC
PC
S
PC
PC
S
PC
PC
PC
PC
PC
PC
PC
PC
PC
PC
PC
S
PC
PC
PC
•. PC • Pulverised Coal b. Based on one compartment out
S • Stoker of
C • Cyclone or
service for cleaning, and one
two out of service, depending
2.1
--
2.13
1.96
2.36
-.
2.03
1.85
2.75
2.75
2.76
1.96
3.35
2.06
3.5
3.">
1.5
6.H6
1 .96
2.6
2.6
1.8?
1.8
2.24
2 0
2.02
?.o
1.75
1.75
1.98
2.72
2.47
2.26
2.1
1.91
1.97
1.97
2.25
2 SI
2.5
2.05
2.42
2.0
2.01
5.12
2.5
2.1
1.77
2.16
2.41
2.31
2.06
2.1
2.43
2.7
-.
1.9
2 .08
2.1
3.4
3.0
2.06
1.81
2.Z
2.9
1.91
2.2
2.2
2.2
2.45
2.45
2.23
2.52
c. RG •
SO •
PJ •
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
SD
RG
SD
SO
SG
PJ
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
SD
RG
SD
RG
RG
PJ
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
RG
SD
SD
RG
RG
RG
SO
RG
RG
RG
RG
RG
KG
RG
RG
Reverse-gas
Shake/Deflate
Pulse Jet
1982
1982
1983
1983
1983/84
1986
1980
1978
1979
1979
1980
1979
1973/74
1983
1979
1979
1975
1978
1964
1979
1979
1979
1982
1982
1986~89
1979
1980
1979
1980
1983
1980
1980
1979
1977
1977
1983
1985
1981
1982/83
1981
1973
1975
1981
1980
1980
1980
1983
1985
1979
1979
1978
1980
1980
1979
1981
1984
1978
1978
1982/84
1978
1980
1979
1983
1981
1978/79
1984
1987
1986
1985
1978
1978
1982
1983/85
So > Stoker w/otl upon the particular case.
B-5
-------
450
400 -
^HH COAL-FIRED CAPACITY
FABRIC FILTER
CONTROL CAPACITY
ACTUAL CAPACITY
PROJECTED CAPACITY
150
100
50
75 76 77 78 79 80 81 82 83 84 85 86 87 88
YEAR*
* YEAR-END TOTALS
89 90
Figure B-2. Actual and projected coal-fired generating capacity
capacity controlled by fabric filters, 1984.1>3~7
and
B-6
-------
require much larger units (in terms of bag dimensions and other design as-
pects), the pulse-jet systems are less effective. Also, because of their
dewaterability in high-temperature, potentially acid environments, coated
fiberglass bags are generally used in utility applications. Since the more-
brittle fiberglass material tends to wear out rapidly when flexed, it is not
suitable for pulse-jet units.
Fabric filters used in utility applications, although similar in basic
design, differ significantly from those used in typical industrial applica-
tions. Utility fabric filters may be 10 to 100 or more times larger than
industrial fabric filters. Because of their larger size and stricter emis-
sion guidelines imposed upon these boilers (even at startup and shutdown),
fabric filters become critical to the operation of the power plant. There-
fore, more attention is directed toward such factors as operation and main-
tenance, energy efficiency, bag life, and preventative maintenance strategies.
Other constraints also affect the design and operation of fabric filters for
utility applications. For example, the temperature of flue gas from utility
boilers is significantly higher than that encountered in many industrial
applications, and the abrasive qualities of the fly ash also must be con-
sidered. The flue gas also contains significant moisture and acid constit-
uents that require high temperatures (above 250°F) to be maintained to pre-
clude acid dewpoint problems and moisture condensation on the bags. If high
temperatures are not maintained, corrosion, bag fabric decay, and bag blinding
can result.
In an effort to maintain the temperatures of the flue gas to the fabric
filters, utilities have installed flange-to-flange insulation. Even with
this insulation, localized corrosion may occur at any heat sinks where sup-
ports and ground-mounted structural beams are welded to the fabric filter
framework. Precautions also must be taken in the "downcomer" sections to the
hoppers. Corrosion can occur on the walls, and the ash may agglomerate in
the hopper as the lower surface temperatures cause condensation. Some low-
sulfur Western coals yield alkaline ashes that tend to "set up" when wetted,
which further complicates the problem of ash removal. Care also must be
taken to prevent inleakage of ambient air in the ash removal system, as this
too reduces the flue gas temperature.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-7
-------
Because fly ash is abrasive, some design feature must be implemented,
particularly at the gas inlet, to minimize the initial impact of the inlet
gas stream on the bag fabric. No standard design is available to ensure
adequate flow distribution of the flue gas (and therefore the fly ash) through
the fabric filter. Some installations have no means of distribution other
than the wedge-shaped inlet manifold; others have baffles, turning vanes, or
a combination of the two.
In some instances, louvered dampers or butterfly valves are used, but
poppet valves at the inlet and outlet of the compartments are most commonly
used for flow control. There may be no real advantage to using poppet valves
at the fabric filter inlet; in fact, there may be a pressure drop penalty.
At the outlet, however, poppet valves prove superior to louver and butterfly
dampers. Poppet valves seal very well. A two-valve design has a lower
pressure drop penalty, because two paths offer less resistance for gas passage.
Using a pair of valves (one large and one much smaller) has the same advan-
tages as multiple gas paths,'but also has the added advantage of reducing bag
stresses during cleaning cycles. Bag reinflation is often accompanied by a
loud "pop" as the flue gas rushes in to fill the void. This damages the
fabric (particularly fiberglass) over a period of time. When the fabric
filter design includes a small (pilot) poppet valve, the reinflation flow can
be started more gradually. The small valve opens first during bag inflation,
and the larger valve opens later to complete the inflation. On reverse-gas
cleaning applications, both economics and a desire to achieve a "gentle"
reinflation dictate the number and size of poppet valves at the outlet.
Either two equal diameter poppet values or one large and one small valve are
commonly used.
Except at the sites of the two pulse-jet installations, woven fiberglass
is the bag material most installations often use. The coatings vary, but
most are Teflon (10 percent by weight). A survey taken in 1981 indicated
that those plants not using Teflon (roughly 13 percent) were using a silicon-
graphite coating or one of the recently introduced acid-resistant finishes.
The woven bags are typically attached to the tube sheet by means of thimbles
8 to 12 inches in height. These thimbles are used to prevent erosion of the
bag material due to fly ash particles entering the bags from the hopper.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-8
-------
As mentioned earlier, utility and industrial fabric filters differ in
several ways. For example, gas volumetric flow rate in electric utility
systems may be as high as 4 x 10 acfm as opposed to 100,000 acfm in typical
industrial applications. Energy costs resulting from ductwork and dust cake
resistance pressure drop are generally much greater for utilities. In addi-
tion, utilities do not benefit from a product recovery credit of the col-
lected material as many industries do. High flue gas temperatures in utility
applications limit the choices of bag fabrics. The volume, flow, tempera-
ture, composition, and particulate concentration of the flue gas entering the
fabric filters in utility applications vary greatly with the boiler load, and
the fly ash represents a wide and often unpredictable range of coal properties.
MONITORING
Utility applications typically incorporate more monitoring devices than
industrial fabric filter systems do to track the operation of the system and
its related equipment. Monitoring and alarm devices display and/or record
the gas flows and pressure losses within the system, incidents involving
compartment isolation, inlet and outlet temperatures of the system, operation
and sequencing of the cleaning apparatus, particulate emissions exiting the
stack, and bag failure (i.e., severe plugging or rupture). Outlet opacity
monitors are typically installed to satisfy environmental regulations, but
they are also useful in detecting problems before they become serious. For
example, when bag rupture problems were encountered at the Harrington Station
of Southwestern Public Service Co., workers were able to pinpoint failures in
the specific compartment through the use of opacity meters.
The outlet opacity monitor should be observed during normal filtering
operation and during compartment cleaning. A gradual increase in opacity
during filtering indicates a worsening bag or compartment leak (assuming the
monitor itself is performing properly). During cleaning, a very clean filter
will show almost no change in opacity as compartments are removed, cleaned,
and put back into service. A drop in opacity when a compartment is removed
from service indicates that the compartment has a leak. The opacity will
also normally increase immediately when that particular module is put back in
service.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-9
-------
The opacity may also increase momentarily when a given compartment is
removed because of the disturbance of an accumulation of ash in the other
compartments resulting from the sudden increase in gas flow in these com-
partments due to the removal of a compartment.
Pressure gauges, level indicators, and gas flow and temperature monitors
also provide data for early detection of problems. Thus, it is apparent that
good monitoring systems, dedicated maintenance, and quality control in the
fabrication and installation of bags lead to greatly improved service and
substantial savings in labor and repair costs.
O&M PROBLEMS AND PRECAUTIONS
Fabric filters have performed well on utility boilers. Design removal
efficiencies for all fabric filters range from 99.4 to 99.9 percent, and in
many cases, actual efficiencies have exceeded the design efficiencies.
Opacities are typically below 5 percent. Pressure drops range between 3 and
12 inches, with newer installations showing values at the lower end of the
range.
Assuming proper fabric filter design and proper bag installation, the
most critical concern is startup and shutdown. Several typical maintenance
problems and precautions are introduced briefly here and illustrated later in
this appendix by case histories.
Operational Factors
Operating factors of concern on fabric filter systems include the clean-
ing system, the bags themselves, the ash-removal system, and overall system
integrity.
Operators must be careful not to clean the bags too frequently. When
bags are cleaned too frequently, the overall average pressure drop is higher
because the dust cake is not as heavy and is harder to remove. Also, fre-
quent bag cleaning weakens the material and shortens bag life.
Operators must minimize the potential for problems associated with
startup and shutdown. If at all possible, the fabric filter system should be
heated thoroughly (e.g., by gas-firing the boiler or by some other means)
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-l 0
-------
before the sulfur-laden gas from coal firing is allowed to enter the collec-
tor. If the fabric filter applied to a coal-fired unit is cold at startup,
moisture from the flue gas will condense on the bags and walls, and the SCU
in the gas will combine with the moisture to form sulfuric acid, which may
result in corrosion and fabric decay. Also, moisture may cause "blinding" of
the fabric when residual fly ash and water seal the air passages in the fiber
weave. During shutdowns and forced outages, fabric filters should be purged
as thoroughly as possible to remove moisture and sulfur-laden gases as the
collector cools.
Operators should observe the performance of the reverse-gas valve and
compressor system to assure adequate bag cleaning during the cleaning cycle.
Operators also must carefully observe the fabric filter monitoring equipment
to detect bag failures as early as possible. A serious bag failure can cause
damage to surrounding bags.
Maintenance Factors
During bag replacement, care must be taken to minimize the risk of
damage to other bags as a result of snags and punctures with tools and equip-
ment. As the utility industry has become more familiar with fabric filter
technology, problems relating to improper maintenance procedures have dimin-
ished in number.
Maintenance personnel must be certain that bag tensioning devices are
properly adjusted and in good condition. One of the primary causes of bag
failure can be traced to improper bag tensioning. Bags also must be installed
properly. Improper installation may cause the bag to rupture and/or become
dislodged. When this occurs, other bags can also be damaged.
Fabric filters are well maintained at most utility applications. The
changeout time for 12-inch-diameter, 36-foot-long fiberglass bags is 15 to 20
minutes (two men). In most fabric filters, insulation is placed between
compartments. Some also have ventilation systems to cool the compartments
quickly, which permits personnel to work comfortably and safely to replace
bags in an isolated compartment while the rest of the fabric filter system is
still in service.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-ll
-------
CASE HISTORIES
The case histories that follow were selected from a population of approx-
imately 84 operating fabric filter installations. Selection was based on the
availability of O&M data and on each case's typicality of U.S. utility fabric
filter installations. Larger installations (500 MW and greater), however,
are not well represented because fabric filters have only recently been
applied on these units. Thus, O&M data are limited. The successes exhibited
with smaller boilers has started a trend by utilities toward equipping larger
plants with fabric filters for particulate emissions control. Some of the
O&M experiences reported herein may become less typical as more is learned
about the design and operation of fabric filter systems en utility boilers.
Several of the references cited herein can be used for further study of U.S.
utility O&M experience with fabric filters.
Colorado Springs Department of Public Utilities, Martin Drake 6
Martin Drake 6 is an 85-MW power generating unit located in Colorado
Springs, Colorado. The boiler is equipped with a reverse-gas design fabric
filter equipped mostly with Teflon B-coated fiberglass bags and a few test
bags with acid-resistant coatings. The coal burned at Martin Drake 6 has a
heating value of 10,200 Btu/lb and moisture, ash, and sulfur contents of 16,
7.5, and 0.37 percent, respectively. The retrofitted fabric filter system
was commissioned into service in 1978.
The system was designed with an air-to-cloth ratio of 2:1 and a flange-
to-flange pressure drop of 4 in. H?0. The unit operates at about 5 inches
pressure drop and has recorded a bag replacement rate of about 1 percent (an
average of less than one bag per month of a total of 2376 bags). Most fail-
ures have occurred between the thimble and the first ring. Some have been
attributed to poor installation and others, to weak spots in the bags. The
reported areas of concern with regard to bags were the clamping devices
and/or procedures used at the thimbles and for bag tensioning.
The utility has experienced problems with temperature instrumentation in
that readings become erratic under certain weather conditions. One possible
solution was to minimize thermocouple junctions and to extend the wiring all
the way to the thermocouple sensor area as much as possible. Whether this
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-l 2
-------
was acted upon is unknown. Other reported problems include general bag
cleaning problems (accompanied by increased pressure drops), tensioning
mechanism problems (such as loss of spring stiffness and ratchet mechanism
wear), and loss of pneumatic control (poppet valve operation) due to cold
weather freeze-up of control air lines. Also reported were sluggish poppet
valve operation on both inlet and outlet, scored cylinders on valve actuators,
and shaft seal problems.
In a recent study at this facility, the residual dust cake weight was
2
about 48 Ib/bag or 0.5 Ib/ft . Overall, the fabric filter system operates
well. The outlet emission rate is 0.005 to 0.006 lb/10 Btu (i.e., a removal
efficiency of 99.93%), which is one of the lowest among U.S. installations.
The utility minimizes the potentially serious problems associated with
startup by firing natural gas. After the fabric filter has been completely
purged with ambient air, it is slowly warmed with the flue gas from the
natural gas firing. Four of the system's 12 compartments are brought on line
at one time; when the entire system is on line, the boiler is switched to
coal-firing.
Kansas City Board of Public Utilities, Kaw 1, 2, and 3
Kaw units 1,2, and 3 (rated at 44, 44, and 68 MW, respectively) are
located in Kansas City, Kansas. All are controlled by reverse-gas fabric
filter systems. The systems on Units 1 and 2 use Teflon B-coated fiberglass
bags whereas the system on Unit 3 uses acid-resistant-coated bags. The fuel
burned at this station is a bituminous coal with a heating value of 11,000
Btu/lb, a moisture content of 6 to 12 percent, an ash content of 15 percent,
and a sulfur content of 5 (max.) percent. Units 1 and 2 were commissioned
into service in 1979; Unit 3 began operation in 1980.
The Kaw fabric filters were designed with air-to-cloth ratios of approxi-
mately 2:1 and flange-to-flange pressure drops of 4 to 6 in. H?0; however,
actual pressure drops fall in the range of 8 to 12 inches, with the average
at the higher end of the range. The high pressure drop is attributed, in
part, to the boiler operations. Occasionally, the boiler has operated in
such a way that the temperature of the gas ducted to the fabric filter has
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-l 3
-------
fallen below dewpoint for extended periods. The resulting moist ash accumula-
tion on the bags reduces bag cleaning effectiveness and yields a high pressure
drop.
Kaw 1 and 2 each experienced about six bag failures per month (1981),
whereas Unit 3 had only three per year. Although bag failures occurred
randomly with respect to bag location, the typical failure was at the rings
on the lower half of the bag itself. Part of the problem at Kaw 1 and 2 is
that the boilers operate in a cycling load; they are not used continuously.
Although subject to a fluctuating load, Kaw 3 is rarely shut down completely.
Heavier-grade bags (13 02.) have since been installed with some success at
Kaw 1 and 2 to minimize the problem. Fan vibration and overall balance also
created some problems, partially because the fans were undersized and par-
tially because of erosion. This could have had an impact on the bag life.
The low-horsepower fan problem was solved by installing larger capacity
units.
Other than the high pressure drop, the primary problem that the utility
reported on Kaw 1 and 2 systems was with reverse-gas fan and fan motor bearing
failures. Both units were designed for removal efficiencies of 99.86 percent,
but efforts to achieve this design efficiency have been unsuccessful. The
units have been unable to achieve the design part.iculate removal efficiencies.
Actual removal efficiencies have been 98.4 percent on Kaw 1 and 98.83 percent
on Kaw 2; outlet dust concentrations have been 0.087 and 0.06 lb/10 Btu,
respectively. This lower removal efficiency is believed to result from the
boiler being a cyclic unit burning a high-sulfur coal with a history of low
flue gas temperatures and from the high pressure drop of the fabric filter.
The removal efficiency of Kaw 1 is one of the lowest recorded among U.S.
utility fabric filters.
Minnesota Power and Light, Clay Boswell 1 and 2
The two 69-MW Clay Boswell power generating units in Cohasset, Minnesota
are owned and operated by Minnesota Power and Light. The fabric filters on
these units are of the reverse-gas design and the bags are woven fiberglass
with Teflon B coating. The boilers burn an 8500-Btu/lb subbituminous coal
that has moisture, ash, and sulfur contents of 25, 10, and 1 percent, re-
spectively. The units began operations in 1979.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-l 4
-------
The fabric filters were designed with an air-to-cloth ratio of 2.26:1
and a flange-to-flange pressure drop of 6 in. hLO. The design particulate
removal efficiency is 99.7 percent. Actual pressure drop is 7 in. H^O and
typical particulate removal efficiency is 99.8 percent.
Individual bag failures do not appear to be a major problem at most
utility installations—one or two bags per month. At Clay Boswell, however,
several hundred bags had to be replaced in one instance as a result of poor
bag tensioning. In 1980, bag failures totaled 100 per year; more recently
the failure has been about six per month. The bag failures usually occur in
the lower 8 feet of the bag. Problems related to boiler tube leaks and low
winter boiler loads in some instances have caused flue gas temperature to
drop below dewpoint for extended periods. The resulting moist ash accumula-
tion on the bags reduced bag cleaning effectiveness and caused a high pressure
drop. When boiler tube leaks were repaired, the problem was eventually
brought under control, and the pressure drop fell back down to 7 in. H?0.
Operator experience appears to have been more instrumental in solving the
pressure drop problem than anything else. High bag failure rate is still a
problem, but no agreement has been reached as to the cause. Flue gas mois-
ture, S03, or a combination of the two in conjunction with the plant's start-
up/shutdown procedures have been suggested as possible causes. The newer
Teflon-core fiberglass bags (used for the past 2 years) have shown a moderate
impoundment in bag life.
Initially, several problems were encountered at the Clay Boswell install-
ations. For example, the units originally failed to meet the design require-
ments of 0.01 grain/scf. After the tube sheet thimbles were seal-welded and
pinhole leaks in the bags were repaired, however, an outlet concentration of
0.007 grain/acf was achieved, which is better than the design requirement.
The utility reported load reductions of 200 hours in 1979 and 250 hours in
1980 due to fabric filter problems.
Reverse-gas fan and fan motor bearing failures also occurred. According
to the log kept on these problems, sluggish poppet valve operation was en-
countered on both the inlet and outlet, cylinders on valve actuators were
scored, and shaft seal problems were noted. In addition, a loss of pneumatic
control (poppet valve operation) resulted from a cold weather freeze-up of
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-l 5
-------
control air lines. Bag tensioning mechanism problems also occurred, such as
loss of spring stiffness and ratchet mechanism wear. The excessive bag
failure mentioned earlier was a direct result of poor bag tensioning that
allowed the bags to droop, which caused creases and wear points.
Finally, problems with the ash handling system were also reported.
Erosion and plugging problems occurred in the vacuum blowers as a result of
ash carryover in the transport.
None of the problems encountered at the Clay Boswell facility proved to
be critical, and operations have improved considerably since startup.
Nebraska Public Power District, Kramer 1, 2, 3, and 4
Units 1, 2, and 3 at the Kramer Power Station in Bellevue, Nebraska, are
rated at 23 MW each; Unit 4 is rated at 36 MW. All the units were started up
in 1977, beginning with Unit 1 in March, and all four were on line by May of
that year. These units represent the first utility fabric filters used at a
plant burning a typical, low-sulfur, Western subbituminous coal in a pulver-
ized coal-fired boiler. The Wyoming subbituminous coal burned at this plant
has a heating value of 10,100 Btu/lb, and moisture, ash, and sulfur contents
of 21, 3.1, and 0.57 percent, respectively. The fabric filter systems are of
reverse-gas design and use typical woven fiberglass bags coated with Teflon
B.
The fabric filters were designed with normal air-to-cloth ratios of
about 2:1 (2.1:1 for Units 1 through 3 and 1.91:1 for Unit 4) and flange-to-
flange pressure drops of about 3 to 5 in. of H?0. Design particulate removal
efficiency of these systems is 99 percent. The units actually have achieved
particulate collection efficiencies of 99.9 percent, outlet emissions of
0.002 lb/10 Btu, opacities of 0.07 percent, and average pressure drops of
4.5 in. H70. The fabric filters on the Kramer units have achieved among the
fi
lowest dust emission concentrations (typically 0.005 to 0.006 lb/10 Btu) of
any U.S. utility fabric filter.
A systematic study of the fabric filter cleaning cycle conducted at
Kramer by Electric Power Research Institute (EPRI) investigators and plant
personnel established for the first time (in commercial operation) that
lengthening the dwell time (time during which no cleaning is taking place in
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS D . ,.
D- ID
-------
any compartment) decreases emissions without affecting average pressure drop.
Further, under some operating conditions, average pressure drop actually de-
creases with increased dwell time. Substituting a 100-minute cleaning cycle
(10-minute dwell time) in place of a 10-minute cleaning cycle (no dwell time)
at Kramer reduced particulate matter penetration 50 percent without increas-
ing the pressure drop. Secondary benefits of less frequent bag cleaning are
reduced stress on the bags, increased equipment reliability, and a lower
average air-to-cloth ratio (as each compartment's time in service versus time
out of service during cleaning was increased).
Bag life exceeds 3 years on the Kramer units. Three of the fabric
filter systems have 10 compartments each; the fourth has 16. Each compart-
ment has 72 bags, for a total of 3312 bags. The bag tension is 50 Ib. Total
bag failures by year were 1 in 1977, 12 in 1978, 28 in 1979, 43 in 1980, and
12 in 1981. An additional 18 test bags failed, which were not included in
the above figures. During the study period, tests were performed with bags
coated with dolomitic lime. As a result, the fabric on Unit 1 experienced
fabric blinding and a pressure drop of more than 10 in. H?0. Fly ash coating
was used thereafter. The bag failures were random with respect to bag loca-
tion in the baghouse, even though the gas distribution is not even across the
compartments.
Other problems have included bearing failures and bent shafts on the
reverse-gas fans and higher-than-design pressure drops. The latter has not
been a big problem, however, because the fans were designed for redundancy.
For boiler startup, the utility has a purge-preheat option. Mechanical
collectors are used along with the fabric filters at boiler startup. First,
gas is fired, and then coal; when the outlet temperature reaches 300°F, the
full gas load is ducted to the fabric filter, and the mechanical collectors,
which were operating parallel to the fabric filter up to this point, are
closed off. Lower power demands in the recent past have necessitated some
cycling of the boilers, and the fabric filters have experienced dewpoint
conditions during these periods. As yet no problems have been reported as a
result of this cycling; overall the utility is satisfied with the performance
of the fabric filters.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-17
-------
Otter Tail Power Co., Coyote 1
Coyote 1 is a 410-MW unit located in Beulah, North Dakota. The boiler
is equipped with a sodium-based spray-dryer FGD system followed by a shake-
deflate design fabric filter. The bags are made of uncoated synthetics
(predominantly acrylic fabric). The coal burned at Coyote is a 7050-Btu/lb
Dakota lignite with moisture, ash, and sulfur contents of 36, 7, and 0.78
percent, respectively. The unit began operations in mid-1981.
The design air-to-cloth (A/C) ratio of the fabric filter is 2.5:1, the
design flange-to-flange pressure drop is 3 to 5 in. H?0, and the design
particulate removal efficiency is 99.5 percent. The actual air-to-cloth
ratio, pressure drop, and removal efficiency were reported to be 3:1, 5 in.
hLO, and 99.53 percent, respectively.
During operation, the flue gas exits two air heaters and flows into a
two-stage flue gas cleaning system for removal of 502 and particulate. This
system consists of four 46-foot-diameter spray dryers that use a sodium
carbonate additive as the SO,, absorbent followed by a 38-compartment fabric
filter. Two axial-flow induced-draft fans discharge the filtered flue gas to
a single stack. The flue gas temperature at the inlet to the fabric filter
is in the range of 210° to 220°F. If flue gas temperatures exceed a pre-
determined setpoint (well within the fabrics capabilities), an alarm is
activated and gas flow is diverted through a fabric filter bypass system.
Although fabric air-to-cloth ratios (i.e., gas flows) have been higher than
anticipated because the boiler uses more excess air than anticipated in the
fabric filter design, the fabric filter has consistently operated well within
expectations. The flue gas flow is also greater because the gas exits the
boiler at a temperature of about 25°F greater than design, which yields a
greater gas volume. Filtration performance (pressure drop, cleanability, and
efficiency) has remained relatively stable in spite of a wide variety of
boiler and spray dryer system operating conditions. Pressure excursions due
to boiler load swings, uneven gas distribution from spray dryers, fabric
filter control, equipment malfunctions, etc., have only been temporary, and
when the system operation returned to normal, so did pressure drop of the
fabric filter.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-l 8
-------
As would be expected, the higher temperature operation (230° to 250°F
compared with the design temperature of 180°F for much of the first 12 months)
resulted in discoloration of the acrylic fabric, but had no serious effect on
fabric strength, expected service life, or dimensional stability. The higher
temperature did accelerate the failure of the polyester fabrics and led to
their replacement with the acrylic material. The utility reported it expects
to do further testing of polyester fabric (at stable operation with tempera-
tures in the 190° to 220°F range) sometime in the future.
Operating and maintenance details were not available on this system.
During the first year-and-a-half of operation the fabric filter underwent a
bag material testing program; therefore, bag replacement rates may not be
meaningful in this case. Reportedly, however, the installation has exhibited
superior performance in terms of low pressure drop at high filter velocities,
fabric replacement experience, and service life expectancy.
Pennsylvania Power and Light, Brunner Island 1
Brunner Island 1 is a 350-MW pulverized-coal-fired unit located in
Yorkhaven, Pennsylvania. The boiler is equipped with a reverse-gas fabric
filter. The bags are made of woven fiberglass with a Teflon coating. Brunner
Island 1 burns an 11,000- to 13,000-Btu/lb eastern bituminous coal with a
moisture content of 5 to 20 percent, an ash content of 12 to 18 percent, and
a sulfur content of 1.1 to 3.0 percent. The unit began operating in October
1980.
The design air-to-cloth ratio is 2.01:1 with two compartments out of
service and 2.21:1 with six compartments out of service. The design pressure
drop is 6 in. H~0, and the design particulate matter removal efficiency is
99.9 percent. The fabric filter system actually operates with a normal pres-
sure drop of only 4 in. H00. The actual outlet emissions generated during
L r
two tests were 0.037 and 0.096 lb/10 Btu vs. the design emission rate cf
0.075 lb/10 Btu. Brunner Island has 24 compartments, with 264 bags per
compartment. The utility reported a total of 6 bag failures in 1980 (the
unit didn't begin operations until October), 209 failures in 1981, and 877 in
1982 (through early December). The large number of failures adversely af-
fected the pressure drop because of the frequent need to cool compartments
for maintenance purposes. The failures occurred randomly throughout the
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-19
-------
system with respect to bag location. The failures of the bags themselves
typically occurred between the thimble and the first ring, and many resulted
from bag cuffing. The tension method caused most of the bags either to be
overtensioned or undertensioned. The utility solved the tensioning problem
by replacing the old stiffer springs with new ones that have superior charac-
teristics. With the old springs, when the bags were ratcheted up a link,
they would be either too tight or too loose and no way was provided to adjust
the tension between the links. The new springs (although very strong) have
better elastic characteristics for this application and yield a more uniform
force on the bags from one link to the next. This factor combined with
ineffective cleaning of the bags resulted in excessive residual dust cakes.
The typical weight of the residual cake on these bags was found to be 1.18
lb/ft2 or 126 Ib per bag.
Problems reported at this facility include those associated with valves
(and valve operators), the control system, and tensioning mechanisms. Boiler
problems such as tube leaks and operating equipment failures have also added
to the problems, as a significant number of load reductions and forced outages
have been reported.
Although many aspects of the Brunner Island fabric filter operation were
initially discouraging, significant improvements have been made lately. Bag
filter total pressure drop, which previously was as high as 12 in. H20 with
all compartments but one in service (one out for bag cleaning), has dropped
to approximately 6 in. H?0 at full load with three compartments out of service
for maintenance and one for bag cleaning. The utility installed horns for
sonic cleaning (eight per compartment) and has been testing several rebagging
strategies (warp in/out, new fabric, etc). Although the bags are still not
satisfactory, the fabric filter operations at the facility have been relative-
ly reliable.
Pennsylvania Power arid Light, Sunbury 1 and 2
The Sunbury units 1 and 2 are located in Shamokin Dam, Pennsylvania.
This station marked the first full-scale fabric filter installation at a
coal-fired generating plant. The combined capacity of the boilers is about
175 MW. Each of the four boilers is controlled by a fabric filter. The coal
blend is 65 to 85 percent anthracite and 15 to 35 percent petroleum coke and
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-20
-------
bituminous coal. The heating value of this blended fuel is about 9800 Btu/lb.
The moisture content is 10 to 20 percent, the ash content is 18 to 30 percent,
and the sulfur content is 1.1 to 1.3 percent.
Preceded by modified mechanical collectors that are approximately 70
percent effective in removing particulate matter, these reverse-gas fabric
filters at Sunbury have a design air-to-cloth ratio of 2.07:1. The bags are
made of woven fiberglass and have a 10 percent by weight Teflon finish. In
operation, the units have demonstrated a particulate matter removal effi-
ciency of 99.9 percent versus a design removal efficiency of 99.2 percent.
Outlet emissions of 0.005 lb/10 Btu were reported, which is among the lowest
reported for U.S. utility fabric filters. The plant reportedly shows no
visible plume, and the average pressure drop has been as low as 3 in. H?0
versus a design pressure drop of 5 in. H?0.
A 4-year bag life is reported at the Sunbury station, the longest at any
U.S. utility installation. The annual bag failure rate is reported to be 4
percent. The residual dust cake weight recorded in a recent study was 0.72
2 12
Ib/ft per bag, or about 68 Ib/bag per day.
Recently, however, concern arose concerning the performance of the
fabric filters when the average flange-to-flange pressure drop rose to about
6 to 6.5 in. O. Although this is cause for concern, the utility has indi-
cated that this is not excessive compared with the pressure drop at other
installations.
Overall, Pennsylvania Power and Light is satisfied with the installation.
Several factors are believed to have contributed to the generally good per-
formance of the Sunbury fabric filters. First, the boilers typically operate
at full load, with minimal swings and unit outages. Second, the bag tension-
ing system at Sunbury permits tensioning at very nearly a steady 50 Ib,
whereas at other installations tensioning may be 50 ± 20 Ib. Other contribut-
ing factors are the use of filter bags that perform well in this specific
environment, a relatively low inlet grain loading, and special gas inlet/out-
let design features. Unlike most fabric filter installations, the flue gas
enters and exits the filter chamber from the center. In typical installa-
tions, the gas enters from the side and exits from the side, and pressure
drops of the bags closest to the inlet may differ as much as an inch from
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-21
-------
those of the bags farthest away. Both the quantity and quality of the ash
collected differ in those two areas. This does not occur at Sunburry where
the more uniform environment is believed to have a beneficial effect on
overall bag life and performance.
Sierra Pacific Power Co., North Valmy 1
North Valmy 1 is a 250-MW power generating station located in North
Valmy, Nevada. The boiler is equipped with a reverse-gas fabric filter in
which Teflon-coated woven fiberglass bags are used. The fuel burned at North
Valmy I is generally a Western low-sulfur coal. The coal originates from
different sources, which accounts for the wide variability of its character-
istics. The heating value ranges from 8000 to 12,250 Btu/lb, and it may have
a moisture content of 3 to 22 percent, an ash content of 3 to 20 percent, and
a sulfur content of 0.3 to 1.5 percent. The unit was commissioned into
service in 1981.
The design air-to-cloth ratio of the fabric filter is 1.99:1 (worst
coal), and the design flange-to-flange pressure drop is 5.5 in. H^O. The
system contains a total of 6480 bags in 10 compartments. The unit is de-
signed for normal operation with eight compartments—one out for cleaning and
one for maintenance. The supplier recommends and the utility practices
filter bag precoating. This coating (fly ash) is applied by slowly placing
the system in service one compartment at a time after 100 percent coal firing
has been achieved.
As of mid-1983, the fabric filter reportedly had not limited boiler
operations. Plant operating procedures contribute to the low bag failure
rate (less than 0.25 percent versus a contract guarantee of 5 percent maximum
failure rate). During the period November 1981 through February 1983, a
total of 24 bags failed.
General problems reported for this installation included those asso-
ciated with valve operation and cold weather. The pneumatic control system
was sometimes subject to temperatures as low as 40°F below zero. Water that
accumulates in the lines would freeze and general moisture/temperature-related
problems caused sluggish operations in equipment throughout the installation
because the pneumatic control system was not designed for this environment.
The utility reported that injecting alcohol in the lines has reduced the
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-22
-------
incidence of freeze-up. Casing and door seal leaking accounted for some
corrosion. Also, unrelated ductwork corrosion and expansion joint failures
have occurred.
On startup, the Nevada Division of Environmental Protection (NDEP)
permits Sierra Pacific to wait until the unit is at approximately half-load
(125 MW), on two pulverizers, before placing the fabric filter in service.
This assures that basically no oil firing is still taking place. This pro-
cedure is based on data developed by the supplier, which strongly suggests
that the bag life will be seriously reduced if oil soot is allowed to accumu-
late on the bags, as would occur during a startup. Sierra Pacific believes
this procedure has greatly contributed to longer bag life.
During a normal unit shutdown, the fabric filter system is kept in
service until the unit is taken off line. The system remains in service for
3 minutes, which allows a hot-air purge of the unit to take place. After the
purge, the unit is suited to the "bypass" mode of operation.
If the outage is to be relatively long (several days), the doors are
opened and the unit's ventilation system is placed in service. If the outage
is expected to be brief, the fabric filter is kept sealed in an effort to
reduce moisture infiltration and heat losses.
The utility is generally satisfied with the performance of the fabric
filter. Except during bag failures, the system continually maintains its
design particulate removal efficiency of 99.7 percent. Opacity readings
typically run 2 to 4 percent; during excursions (bag ruptures) the opacity
readings rise to the 8 to 10 percent range.
Southwestern Public Service Co., Harrington 2 and 3
Harrington units 2 and 3 are located in Amarillo, Texas. Each is rated
at 350-MW. The fabric filter system on Harrington 2 was the first large unit
installed on a new utility boiler. The fabric filters on both units are of
the shake-deflate design. Originally equipped with silicon/graphite-coated
woven fiberglass bags, these units were later switched to Teflon B woven
fiberglass bags. The units are fired with an 8230 Btu/lb Western subbituminous
coal with moisture, ash, and sulfur contents of 30, 6.4, and 0.48 percent,
respectively. Harrington 2 began operations in mid-1978, and Harrington 3
started up in 1980. The design air-to-cloth ratio of the fabric filter on
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-23
-------
Unit 2 is 3.4:1, and the design pressure drop is 6 in. H^O. The design air-
to-cloth ratio for the Number 3 unit is 3.0:1, and the design pressure drop
is 7 in. H00. Actual pressure drop on this unit is 6 in. H,,0. Pressure
drops on both units have reached as high as 13 in. I-LO. The pressure drop
now is kept below 10 inches, primarily as a result of studies the utility has
performed to optimize bag shaking and overall cleaning cycles. The actual
particulate removal efficiency for both units is reported to be 99.7 percent
for these units with outlet emission concentrations of 0.02 lb/10 Btu.
The Harrington facility has hosted fabric filter studies OP bag fabrics,
effects of shaking frequency on pressure drop, and other subjects of concern.
As a result of such studies, more suitable bags were identified (the utility
now uses 10-oz Teflon-coated fiberglass or 14-oz Teflon- or acid resistant-
coated fiberglass bags, the frequency of shaking was increased, and the
shaker support mechanism was redesigned. The studies demonstrated the viabil-
ity cf using shake-deflate fabric filters for large-scale installations.
No fabric-filter-related problems have been reported during startup.
The startup procedure at Harrington begins with gas firing to bring the
boiler outlet gas temperature to about 250°F. The fabric filter is then
heated, end when fully on line, the boiler fuel source is switched to coal.
The utility has reported no problems traceable to startup or shutdown at
Harrington.
Southwestern Public Service Co. is generally satisfied with the opera-
tions of the Harrington fabric filters. The utility would like to see im-
provements made in bag life. Its experience has been 3 to 3.5 years on the
best bags. Some studies seem to imply that increasing the filtering time
before cleaning could extend the bag life an additional year.
Texas Utilities Co., Monticello 1 and 2
The Monticello steam electric station is located in Mt. Pleasant, Texas.
Units 1 and 2 are each rated at 575 MW. The boilers are controlled by fabric
filters of the shake-deflate design and parallel electrostatic precipitators.
The filter bags are Teflon B-coated woven fiberglass. The coal burned at
Monticello is a 5750- to 8000-Btu/lb Texas lignite with a moisture content of
26 to 37 percent, an ash content of 5.8 to 23 percent, and a sulfur content
of 0.3 to 2.03 percent. The units were commissioned in 1978 and 1979.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-24
-------
Each of the retrofitted fabric filter systems has a design ratio of
2.9:1 and a flange-to-flange pressure drop of 9 in. H^O. The actual pressure
drop has ranged from 9 to 11 in. H00, but recently was reported to be 12
inches. The utility reported that the pressure drop is a function of ash
loading and the nature of the fabric filter system. The units now operate in
the 10- to 12-inch range. The facility is kept below 12 inches through care-
ful attention to the bag cleaning operations. The residual dust cake weight
i
was measured at 0.33 1b/ftc or about 33 Ib per bag. Opacity readings range
from about 4 to 20 percent.
The utility originally used fiberglass bags coated with a silicon-graphite
material, but massive failures occurred after only 4 to 6 months of operation.
The primary cause of the bag failures was bag tensioning and subsequent weave
tightening; it was impossible to maintain the proper bag tension. These bags
accumulated as much as 100 Ib of residual ash. Later, the Teflon B-coated
fiberglass bags were installed, and many of these have lasted more than 2
years.
The fabric filters were originally designed to control about 80 percent
of the gas flow, and the parallel electrostatic precipitators were to control
the other 20 percent. Because of problems associated with having these two
control devices in parallel, dedicated fabric filter fans had to be installed
to maintain the gas flow to these systems.
Pressure drop and relatively short bag life are the two primary concerns
voiced by the utility because of the substantial costs associated with these
problems. Each fabric filter contains 7344 bags; one unit has been com-
pletely rebagged twice, the other has been rebagged once. The utility
believes, however, that the 2-year bag life achieved only on the best bags in
the past has now become the expected bag life overall.
Recently, the utility has achieved pressure drops as low as 10 to 10.5
in. HpO on the fabric filters, but no further improvements have been noted
since then. The inlet grain loading on these fabric filters is relatively
high (9 to 10 grains/acf), and the 12-inch pressure drop mentioned earlier is
one of the highest ever reported for any U.S. utility fabric filter installa-
tion.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-25
-------
Other general problems reported have included casing and door seal
leaking, which resulted in corrosion and flue gas distribution problems.
Although the poor gas distribution has not caused serious O&M problems,
heavier ash loading has been noted in the hoppers beneath some compartments,
which is not expected because the inlet ducts have gas distribution devices
built into them. Failures of the reverse gas fans and fan motors have also
been noted, and pressure blower erosion in the ash handling system has re-
sulted from ash entrainment in the incoming air stream. These problems are
controlled through operation and maintenance measures rather than major
design modifications. Some indications of improved availability are evident
for the Monticello units. Reported restricted hours for these units are as
follows: 740 in 1978, 144 in 1979, and 22 in 1980.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-26
-------
REFERENCES
1. Carr, R. C., and W. B. Smith. Fabric Filter Technology for Utility
Coal-Fired Power Plants. J. of the Air Pollution Control Association,
31(2):178-181, 1984.
2. Perkins, R. P. Industrial Fabric Filter Experience Applicable to Utilities.
In Proceedings of First Conference on Fabric Filter Technology for Coal-
Fired Power Plants, Denver Colorado, July 15-17, 1981. CS-2238. p.
3-37.
3. U.S. Department of Energy. Inventory of Power Plants in the United
States. Office of Utility Project Operations. DOE/RA-0001, December
1977.
4. U.S. Department of Energy. Inventory of Power Plants in the United
States - December 1979. Energy Information Administration. DOE/EIA-
0095(79), June 1980.
5. U.S. Department of Energy. Inventory of Power Plants in the United
States - 1980 Annual. Energy Information Administration. DOE/EIA-0095(80),
June 1981.
6. U.S. Department of Energy. Inventory of Power Plants in the United
States - 1981 Annual. Energy Information Administration Office of Coal,
Nuclear, Electric, and Alternate Fuels. DOE/EIA-0095(81), September
1982.
7. Personal communication from Mr. Skeer, Office of Policy Planning and
Analysis, U.S. Department of Energy, August 1983.
8. Piulle, W., and R. Carr. Operating History and Current Status of Fabric
Filters in the Utility Industry. In Proceedings of First Conference on
Fabric Filter Technology for Coal-Fired Power Plants, Denver, Colorado,
July 15-17, 1981. CS-2238. p. 1-5.
9. Piulle, W., and R. Carr. 1983 Update, Operating History and Current
Status of Fabric Filters in the Utility Industry. In: Proceedings of
Second Conference on Fabric Filter Technology for Coal-Fired Power Plants
Denver, Colorado, March 22-24, 1983. EPRI CS-3257. p. 1-5.
10. Carr, R. C., and W. B. Smith. Fabric Filter Technology for Utility
Coal-Fired Power Plants. J. of the Air Pollution Control Association
31(2) and 31(3), 1984.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS
B-27
-------
REFERENCES
11. Electrical Power Research Institute (EPRI). Proceedings of First Con-
ference on Fabric Filter Technology for Coal-Fired Power Plants. EPRI
CS-2238, February 1982.
12. EPRI. Proceedings of Second Conference on Fabric Filter Technology for
Coal-Fired Power Plants. EPRI CS-3257, November 1983.
APPENDIX B-OPERATION AND MAINTENANCE OF UTILITY FABRIC FILTERS B-28
-------
GLOSSARY OF TERMINOLOGY
ABRASION - FLEX: Where the cloth has abraded in
a creased area by repeated bending.
ABRASION - SURFACE: Where the cloth surface has
been abraded by rubbing, scuffing, erosion.
ABSOLUTE ZERO: The zero from which absolute
temperature is reckoned. Minus 460°F., approxi-
mately.
ACETATE: A manufactured fiber in which the fiber
forming substance is cellulose acetate.
ACRYLIC: A man-made polymerized fiber which con-
tains at least 85% acrylonitrile.
AEROSOL: An assemblage of small particles, solid
or liquid, suspended in air or gas.
AIR, DRY: In psychrometry, air containing no water
vapor.
AIR, STANDARD: Air with a density of 0.075 Ib. per
cubic foot. This is substantially equivalent to dry
air at 70°F. and 29.92 in. (Hg) barometer.
AIR-TO-CLOTH RATIO: The volumetric rate of capacity
of a fabric filter; the volume of air (gas) cubic
feet per minute, per square foot of filter media
(fabric).
ANEMOMETER: An instrument for measuring the
velocity of air or gas.
ATMOSPHERIC PRESSURE: The pressure of the
atmosphere as measured by means of the
barometer at the location specified.
BACKWASH: A method of fabric cleaning where direc-
tion of filter flow is reversed, accompanied by flexing
of the fabric and breaking of the dust cake. Also
known as backpressure, repressure, collapse-clean,
etc.
BAG: The customary form of filter element. Also known
as tube, stocking, etc. Can be unsupported (dust
on inside) or used on the outside of a grid support
(dust on the outside).
BATCH CLEANED: Usually refers to a process used
in heat cleaning fiber glass cloth in roll form by
exposing it at 500°F. to 600°F. for prolonged
periods to burn off the starches or binders.
BLAST GATE: A sliding plate installed in a supply
or exhaust duct at right angles to the duct for the
purpose of regulating air flow.
BLINDING (BLINDED): The loading, or accumulation,
of filter cake to the point where capacity rate is
diminished. Also termed "plugged".
BRITISH THERMAL UNIT (btu): The amount of
heat required to raise one pound of water one de-
gree fahrenheit.
BROKEN TWILL: Modified twill weave where the
diagonal twill line is shifted in a regular pattern.
BULKED YARN: Multi-filament yarn which has been
processed by high pressure air passing through the
yarn and relaxing it into gentle loops, bends, etc.
CALENDERING: The application of either hot or cole"
pressure rolls to smooth or polish a fabric, there
by reducing the thickness of the cloth and decreas-
ing air permeability.
CANTON FLANNEL: Usually a twill weave fabric with
the filling float heavily napped.
CHAIN WEAVE: A 2/2 brofcen twill weave, arranged
2 threads right and 2 left
CLOTH: In general, a pliant fabric; - woven, knitted,
felted, or otherwise formed of any textile fiber,
wire, or other suitable material. Usually understood
to mean a woven textile fabric.
CLOTH WEIGHT: Is usually expressed in ounces per
square yard or ounces per square foot. However,
cotton sateen is often specified at a certain number
of linear yards per pound of designated width. For
example, a 54" - 1.05 sateen weighs 1.05 linear
yards per pound in a 54" width.
CONDENSATION: The process of changing a vapor
into liquid by the extraction of heat.
CORONIZING: A heat cleaning process for fiber glass
fabric to burn off the starches (used in processing)
usually at temperatures of 1000 F. for short duration.
CORROSION: Deterioration or physical degredationdue
to chemical action.
COTTON NUMBER: Staple yarns are generally sized
on the cotton system. Example: an 18 singles yarn
is of such size that 18 hanks (each hank contains
840 yards) weighs one pound.
COUNT: The number of warp yarns (ends) and fillu
yarns (picks) per inch. Also called thread count.
COVER: A description term for the appearance of
woven goods. A well covered cloth is the opposite
of an open, or "reedy" cloth.
CRIMP: The corrugations in a yarn from passing over
and under other yarns at right angles.
CROWFOOT SATIN: A 3/1 broken twill arranged 2
threads right, then 2 threads left, etc. Also called
4 shaft satin, or broken crow weave.
DAMPER: An adjustable plate installed in a duct
for the purpose of regulating air flow.
DEHUMIDIFY: To reduce by any process the quantity
of water vapor.
DENIER: The number, in grams, of a quantity of yarn,
measuring 9000 meters in length. Example:
A 200 denier yarn measuring 9000 meters weighs
200 grams. A 200/80 yarn indicates a 200 denier
yarn composed of 80 filaments. Usually used for
continuous multi-filament yarns of silk, rayon, Orion?
Dacron*Dynel« Nylon'etc.
DENSITY: The ratio of the mass of a specimen of a
substance to the volume of the specimen. The mgss
of a unit volume of a substance. Dry air at 70 F.
and 29.92" Hg has a density of 0.075 pounds per
GLOSSARY OF TERMINOLOGY
6-1
-------
cubic foot.
DIMENSIONAL STABILITY: Ability of the fabric to
retain finished length and width, under stress, in
hot or moist atmosphere.
DRILL: Same as twill except the diagonal twill line
usually runs from lower right to upper left. A 2/1
LH twill, or 3/1 twill.
DUST: Solid particles less than 100 microns created
by the attrition of larger particles. Particles thus
formed are not usually called dust unless they are
larger than about 1 micron diameter.
DUST COLLECTOR: A device to remove solid aerosol
particles from a gas stream.
DUST LOADING: The weight of solid paniculate
suspended in an air (gas) stream, usually expressed
in terms of grains per cubic foot, grams per
cubic meter or pounds per thousand pounds of gas.
DUST PERMEABILITY: Defined as the mass of dust
(grains) per square foot of media divided by the
resistance (pressure drop) inches w.g. per unit of
filtering velocity, fpm. Not to be compared with
cloth permeability.
END: An individual yarn or cord; a warp yarn running
lengthwise of the fabric.
ENTRY LOSS: Loss in total pressure caused by air
(gas), flowing into a duct or hood (usually expressed
in inches w.g.).
ENVELOPE: A common form of filter element.
EROSION: Wearing away due to mechanical action.
EXTENSIBLILITY: The stretching characteristic of
fabric under specific conditions of load, etc.
FABRIC: A planar structure produced by interlacing
yarns, fibers or filaments.
KNITTED Fabrics are produced by interlocking
strands of yarn, etc.
WOVEN Fabrics are produced by interlacing
strands at more or less right angles.
BONDED Fabrics are a web of fibers held
together with a cementing medium which does not
form a continuous sheet of adhesive material.
FELTED Fabrics are structures built up by the
interlocking action of the fibers themselves, without
spinning, weaving or knitting.
FIBER: The fundamental unit comprising a textile
raw material such as cotton, wool, etc.
FILAMENT: A continuous fiber.
FILL: Crosswise threads woven by loom.
FILL COUNT: Number of fill threads per inch of cloth.
FILLING YARN: Yarns in a fabric running across the
width of a fabric; i.e., at right angles to the warp.
FILTER DRAG: Pressure drop, inches w.g. per cubic
foot of air per minute, per square foot of filter media.
Analogous to the resistance of an element in an
electrical circuit. The ratio of filter pressure
to filter velocity.
FILTER MEDIA: The substrate support for the filter
cake; the fabric upon which the filter cake is built.
FILTER VELOCITY: The velocity, feet per minute,
at which the air (gas) passes through the filter
media, or rather the velocity of approach to the
media. The filter capacity rate.
FILTRATION RATE: The volume of air (gas), cubic
feet per minute, passing through one square foot of
filter media.
FINISHED: A fabric which has been processed after
weaving, i.e., other than in the greige.
FLAME RETARDANT: A finish designed to repel
the combustibility of a fabric, either of a durable or
non-durable type.
FLOAT: The position of a yarn that passes over two
or more yarns passing in the opposite direction.
Example: in standard cotton sateen, yarns "float"
four, and pass under one. In other words 4/1.
FLUOROCARBON: Fiber formed of long chain carbon
molecules, available bonds saturated with fluorine.
FOG: Suspended liquid droplets generated by condensa-
tion from the gaseous to the liquid state, or by
breaking up a liquid into a dispersed state, such as
by splashing, foaming and atomizing. (See mist)
FULLED: A woven fabric treated to raise fiber ends
(like napping) so that the thready, woven look is
partially or completely obscured.
FUME: Fine particles dispersed in air or gases,
formed by condensation, sublimation or chemical
reaction. Particles are usually less than one micron
in size.
GAS: Gas is a formless state of matter completely
occupying any space. Air is a gas.
GLASS (FIBER-GLASS): A manufactured fiber in which
the fiber forming substance is glass.
"GRAB" TENSILE: The tensile strength, in pounds
per inch, of a textile sample cut 4" x 6" and pulled
in two lengthwise by two 1" square clamp jaws
set 3" apart and pulled at a constant specified
speed.
GRAIN: 1/7000 pound or approximately 65 milli-
grams.
GRAVITY, SPECIFIC: The ratio of the mass of a unit
volume of a substance to the mass of the same volume
of a standard substance at a standard temperature.
Water is usually taken as a standard substance.
For gases, dry air at the same temperature and
pressure as the gas is often taken as the standard
substance.
GREIGE CLOTH: Cloth as it comes off the loom,
or so-called "loom finish".
GRID CLOTH: The cloth used in supporting the sliver
in making a supported, needled felt.
HAND OR HANDLE: The "feel" of the cloth - as soft,
harsh, smooth, rough, silken-like, boardy, etc.
GLOSSARY OF TERMINOLOGY
G-2
-------
HARNESS: The frame used to raise or lower those
warp yarns necessary to produce a specific weave
at the same time permitting the filling to be passed
through by the shuttle.
HEAD END: A piece of fabric taken from the end of
a roll of cloth.
HEAD SET: A finishing process for that particular
fiber, in fabric form, to stabilize it against
further shrinkage at predetermined temperatures.
HEAT, SPECIFIC: The heat absorbed (or given up)
by a unit mass of a substance when its temperature
is increased (or decreased) by one degree. The
common unit is the btu per degree Fahrenheit.
For gases, both specific heat at constant pressure
(c ) and specific heat at constant volume (c ) are
friquently used.
HOOD SUCTION: The entry loss plus the velocity
pressure m the connecting duct
HUMIDITY, ABSOLUTE: The weight of water vapor
carried by a unit weight of dry air or gas. Pounds
of water vapor per pound of dry air; grains of water
vapor per pound of dry air.
HUMIDITY, RELATIVE: The ratio of the absolute
humidity in a gas to the absolute humidity of a
saturated gas at the same temperature.
HYDROPHILIC FIBERS: Those fibers not readily water
absorbent.
HYGROSCOPIC: Those fibers which are water
absorbant-
INCH OF WATER: A unit of pressure equal to the
pressure exerted by a column of liquid water one
inch high at a standard temperature. The standard
temperature is ordinarily taken as 70 F. One inch
of water at 10 F. = 5.196 Ib per sq. ft.
INTERLACING: The points of contact between the warp
and filling yarns in a fabric.
INTERSTICES: The openings between the interlacings
of the warp and filling yarns; i.e., the voids.
K FACTOR: The specific resistance of the dust cake,
inches water gage per pound of dust per square
foot of filter area per feet per minute filtering
velocity.
LOOM FINISH: Same as greige cloth.
MANOMETER: An instrument for measuring pressure;
a U-tube partially filled with a liquid, usually water,
mercury or a light oil, so constructed that the
amount of displacment of the liquid indicates the
pressure being exerted on the instrument.
MICRON: A unit of length, the thousandth part of 1 mm
or the millionth of a meter, (approximately 1/25,000
of an inch).
MILDEW RESIST FINISH: An organic or inorganic
finish to repel the growth of fungi on natural fibers.
MIST AND FOG: A distinction sometimes made be-
tween mist and fog is of minor importance since both
terms are used to indicate the paniculate state
of airborne liquids. Mist is a visible emission
usually formed by a condensation process or a vapor-
phase reaction, the liquid particles being sufficient)
large to fall of their own weight.
MODACRYLIC: A man-made fiber which contains
less than 85% acrylonitrile (at least 35%).
MOL: A weight of a substance numerically equal to
its molecular weight. 11 the weight is in pounds,
the unit is "Pound Mol". For dry air at 70 F.,
and a pressure of one atmosphere, a pound mol
occupies 386 cubic feet.
MONOFILAMENT: A continuous fiber of sufficient
size to serve as yarn in normal textile operations.
MULLEN BURST: The pressure necessary to rupture
a secured fabric specimen, usually expressed in
pounds per square inch.
MULTIFILAMENT: (Multifil) A yarn bundle composed
of a number of filaments.
NAPPED: A process to raise fiber or filament ends
(for better coverage and more surface area) ac-
complished by passing the cloth over a large re-
volving cage or drum of small power-driven rolls
covered with card clothing (similar to a wire brush).
NEEDLED FELT: A felt made by the placement of
loose fiber in a systematic alignment, with barbed
needles moving up and down, pushing and pulling
the fibers to form an interlocking of adjacent fibers.
NON-WOVEN FELT: A felt made either by needling,
matting of fibers or compressed with a bonding agent
for permanency.
NYLON: A manufactured fiber in which the fib.'
forming substance is any long-chain synthetic poly-
amide having recurring amide groups.
OLE FIN: A manufactured fiber in which the fiber form-
ing substance is any long-chain synthetic polymer
composed of at least 85% by weight of ethylene,
propylene, or other olefin units.
PERMEABILITY, FABRIC: Measured on Frazier
porosity meter, or Gurley permeometer, etc. Not
to be confused with dust permeability. The ability
of air (gas) to pass through the fabric, expressed in
cubic feet of air per minute per square foot of fabric
with an 0.5" HO pressure differential.
PICK: An individual filling yarn running the width
of a woven fabric at right angles to the warp. In
England it is termed woof, or weft
PICK GLASS: A magnifying glass used in counting
the warp and filling yarn in the fabric.
PITOT TUBE: A means of measuring velocity pressure.
A device consisting of two tubes - one serving to
measure the total or impact pressure existing in
an air stream, the other to measure the static
pressure only. When both tubes are connected across
a differential pressure measuring device, the static
pressure is compensated automatically and the
velocity pressure only is registered.
GLOSSARY OF TERMINOLOGY
6-3
-------
PLAIN WEAVE: Each warp yarn passing alternately
over each filling yarn. The simplest weave, 1/1
construction. Also called taffeta weave.
PLENUM CHAMBER: An air compartment maintained
under pressure, and connected to one or more ducts.
A pressure equalizing chamber.
PLY: Two or more yarns joined together by twisting.
POLYESTER: A manufactured fiber in which the fiber
forming substance is any long-chain synthetic
polymer composed of at least 85% by weight of an
ester of dihydric alcohol and terephthalic acid.
POROSITY, FABRIC: Term often used interchangeably
with permeability. Actually percentage of voids
per unit volume - therefore, the term is improperly
used where permeability is intended.
PRESHRUNK: Usually a hot aqueous immersion of
the cloth to eliminate its tendency to shrink in
further wet performances.
PRESSURE, ATMOSPHERIC: The pressure due to the
weight of the atmosphere, as indicated by a
barometer. Standard atmospheric pressure is 29.92"
of mercury equivalents in other units are 760 mm
of mercury, 14.7 psia, and 407 inches water column.
PRESSURE, GAGE: Pressure measured from
atmospheric pressure as a base. Gage pressure may
be indicated by a manometer which has one leg
connected to the pressure source and the other
exposed to atmospheric pressure.
PRESSURE JET CLEANING: A bag cleaning method
where a momentary burst of compressed air is
introduced through a tube or nozzle attached to the top
cap of a bag. A bubble of air flows down the bag,
causing bag walls to collapse behind it.
PRESSURE, RESISTANCE: Resistance pressure (RP)
is the pressure required to overcome the resist-
ance of the system. It includes the resistance of
straight runs of pipe, entrance to headers, bends,
elbows, orifice loss, and cleaning device. It is
indicated by the difference of total pressure between
two points in the duct system.
PRESSURE, STATIC: The postential pressure exerted
in all directions by a fluid at rest. For a fluid in
motion, it is measured in a direction normal to the
direction of flow. Usually expressed in inches water
gage, when dealing with air.
PRESSURE, TOTAL: The algebraic sum of the velocity
pressure and the static pressure (with due regard
to sign). In gas-handling systems these pressures
are usually expressed in inches water gage. The
sum of the static pressure and the velocity pressure.
PRESSURE, VELOCITY: The kinetic pressure in the
direction of flow necessary to cause a fluid at rest
to flow at a given velocity. Usually expressed in
inches water gage.
PULSE JET: A system of bag cleaning using a
momentary burst of compressed air in the discharge
nozzle of a filter bag, which stops filter flow and
inflates the bag in the opposite direction.
RAVEL STRIP TENSILE: The tensile strength, in pounds
per inch of a 6" long textile sample cut just over
one inch wide, (with yarns peeled off each side
down to exactly one inch wide) pulled in two length-
wise between jaws set 3" apart and pulled at a
constant specified speed.
RAYON: A manufactured fiber composed of regenerated
cellulose.
REED MARKS: The identations between 2, 3 or 4 ends,
usually eliminated in finishing.
REPEAT: The number of threads in a weave before
the weave repeats or starts over again. The number
of ends and picks in the repeat may be equal or
unequal, but in every case the repeat must be in
a rectangular form.
RESISTANCE: Analogous to electrical resistance -
the pressure drop across the filter media and dust
cake, expressed in inches water gage.
REVERSE JET CLEANING: A cleaning method
(Mersey) using a traveling ring traversing the ex-
terior of the filter bag. High pressure air is blown
backwards through the fabric through small holes
or slots in contact with the cloth.
SANFORIZED: A patented process where the cloth
is "puckered" in the warp direction to eliminate
shrinkage in laundering.
SARAN: Any long-chain synthetic polymer composed
of at least 85% vmylidene - chloride units.
SATEEN: Cotton cloth made with a 4/1 satin weave,
either as warp sateen or filling sateen.
SATIN WEAVL: A fabric usually characterized by
smoothness and luster. Generally made warp face
with a great many more ends than picks. The sur-
face consists almost entirely of warp (or filling)
floats in construction 4/1 to 7/1. The intersection
points do not fall in regular lines, but are shifted
in a regular or irregular manner-
SCOUR: A soap and hot water wash to "off loom"
fabric.
SELVAGE: The binding lengthwise edge of a woven
fabric.
SHAKING (CLEANING): A common, mechanical method
of removing dust from filter elements. Backwash,
or other supplemental methods, are often used with
shaking.
Air-shaking is a bag cleaning means wherein bags are
shaken in a random fashion by high velocity air
stream ratner than by mechanical devices.
SINGEING: The burning off of the protruding hairs
from the warp and filling yarns of the fabric.
SINGLES: The term used to imply only one yarn.
SIZING: A protective coating applied to yarn to insure
safe handling; i.e., abrasion-resistance during
weaving.
GLOSSARY OF TERMINOLOGY
G-4
-------
SLEAZY: Lacking in firmness or substance; thin,
flimsy.
SLIPPAGE: The movement or shilling of yarns in a
fabric from their normal position.
SLUB: A. heavy accumulation of fiber or lint carried
on a yarn and interlocked during weaving.
SMOKE: An air suspension (aerosol) of particles,
usually but not necessarily solid, originating from
a combustion process.
SONIC (SOUND): A fabric cleaning method using acoustic
energy to vibrate the filter elements. Used alone,
or as a supplement to shaking, or backwash cleaning.
SPUN FABRIC: Fabric woven from staple (spun)
fiber - same as staple.
STANDARD ATMOSPHERE: The pressure exerted by
a column of mercury 29.92" high at 70°F approx-
imately 14.7 psi.
STAPLE FIBER: Man-made fibers cut to specific
length - 1-1/2", 2", 2-1/4" etc. - natural fibers
of a length characteristic of fiber, animal fibers
being the longest;.
"S" TWIST: The yarn spirals conform in slope to
the center portion of the letter "S".
TAFFETA: Closely woven plain weave (1/1) fabrics
with the warp yarns greatly outnumbering the filling
yarns.
TEMPERATURE, ABSOLUTE: Temperature expressed
in degrees above absolute zero.
TEMPERATURE, DEW-POINT: The temperature at
which the condensation of water vapor in a space
begins for a given state of humidity and pressure
as the temperature of the vapor is reduced. The
temperature corresponding to saturation (100 per
cent relative humidity) for a given absolute humidity
at constant pressure.
TEMPERATURE, DRY-BLUB: The temperature of a
gas or mixture of gases indicated by an accurate
thermomenter after correction for radiation.
TEMPERATURE SCALES: Temperature scales, Centi-
grade and Fahrenheit derive their degree values
by dividing the difference between the ice point and
steam points of water as follows: Centigrade 100
and Fahrenheit ISO. The value of a Fahrenheit degree
is therefore 5/9 of a Centigrade degree. The Fahr-
enheit scale is generally used in air handling practice.
The Rankine scale, sometimes called Fahrenheit
absolute, has its zero at the lowest attainable
temperature, exactly 459.61 degrees below the zero
of the Fahrenheit scale. To convert Fahrenheit to
Rankine temperature (generally designated R), add
459.67 degrees, (460 is sufficiently accurate).
TEMPERATURE, WET-BULB: Wet bulb temperature
is a measure of the moisture content of air (gaa).
It is the temperature indicated by a wet bulb
psychrometer.
TENACITY: Ultimate tensile strength of a fiber,
filament, yarn, etc. expressed in grama per denier
(g.p.d.).
TENSILE STRENGTH: The ability of yam or fabrj
to resist breaking by direct tension. Ultimate beai,
ing strength in psi.
TENTER FRAME: (Pin tenter) A machine for drying
cloth under tension. Tentering. Also called framing.
TEXTILE: That which is or may be woven. Comes
from the Latin "Texere", to weave. Hence any kind
of fabric.
THREAD COUNT: The number of ends and picks per
inch of a woven cloth. For example 64x60 (ends
count first).
THROW: Process of doubling or twisting fibers into
a yarn of the desired size and twist.
TOW: A large number of filaments collected in a
loose rope-like form, without definite twist.
T.P.I.: Twist per inch. (Turns per inch).
TWILL WEAVE: Warp yarns floating over or under
at least two consecutive picks from lower to upper
right, with the point of intersection moving one
yarn outward and upward or downward on succeeding
picks, causing diagonal lines in the cloth.
TWIST: The number of complete spiral turns per inch
in a yarn, in a right or left direction; i.e., "S"
or "Z" respectively.
VAPOR: The gaseous form of substances which are
normally in the solid or liquid state and which can
be changed to these states either by increasing the
pressure or decreasing the temperature.
VELOCITY HEAD: Same as velocity pressure. (S
Pressure, Velocity).
VELOCITY OF APPROACH: The velocity of air (gas),
feet per minute, normal to the face of the filter
media.
VELOCITY TRAVERSE: A method of determining the
average air velocity in a duct. A duct, round or
rectangular, is divided into numerous sections of
equal area. The velocity is determined in each area
and the mean is taken of the sum.
VOLUME, SPECIFIC: The volume of a substance per
unit mass; the reciprocal of density; usually given
in cubic feet per pound, etc.
WARP: Lengthwise threads in loom or cloth.
WARP BEAM: Large spool-like or barrel-like device
on which the warp threads are wound.
WARP COUNT: Number of warp threads per inch of
width.
WARP SATEEN: The face of the cloth having the
warp yarns floating over the filling yarns and
being greater in number than the filling yarns
(per inch).
WEAVE: The pattern of weaving; i.e., plain, twill,
satin, etc.
WEFT: Same as filling, the crosswise threads (yams).
WOOF: Same as filling or weft
WOOLEN SYSTEM: A "system" of yarn manufacturing
for spinning wool fiber into yarn, usually more open
and not aligned as parallel as the cotton system.
WORSTED SYSTEM: A system of yarn namufacturing
suited for medium and longer wools. Includes ad-
ditional processing steps resulting in the most uniform
yarn. The resulting yarn is compact and level.
WOVEN FELT: Predominantly a woven woolen fabric
heavily fulled or shrunk with the weave being com-
pletely hidden due to the entanglement of the woolen
fibers.
YARN: Twisted fibers or filaments in a continuous
strand suitable for weaving, etc. Ply yarn is formed
by twisting two or more single yarns together. Ply
yarns are in turn twisted together to form cord.
YARN SIZE (DENIER, OH COUNT): A relative measure
of fineness or coarseness of yarn. The smaller the
number in spun yarns, the coarser the yarn. The
higher the denier of a filament yarn, the coarser
(heavier) the yarn.
"Z" TWIST: The yarn spirals conform in slope to the
center portion of the letter "Z".
GLOSSARY OF TERMINOLOGY
6-5
-------
REFERENCE
Industrial Gas Cleaning Institute (IGCI). Fundamentals of Fabric Col-
lectors and Glossary of Terms. Publication F-2. Stamford, Conn. 1972,
GLOSSARY OF TERMINOLOGY
G-6
»U.S. Government Printing office: 1988—548-158/67134
------- |