United States
                Environmental Protection
                Agency
               Office of Air Quality
               Planning and Standards
               Research Triangle Park NC 27711
EPA-34O/1-83-O17
January 1983
                Stationary Source Compliance Series
vvEPA
Kraft  Pulp Milj
Inspection Guide

-------

-------
                                EPA-340/1-83-017
    Kraft Pulp Mill
  Inspection  Guide
                by

       PEDCo Environmental, Inc.
         11499 Chester Road
         Post Off ice Box 46100
      Cincinnati, Ohio  45246-0100
       Contract No. 68-01-6310
       Work Assignment No. 65
           PN 3 660-1-65
      John R: Busik, Project Officer
    Robert C. Marshall, Task Manager
U.S. ENVIRONMENTAL PROTECTION AGENCY
   Stationary Source Compliance Division
         401 M Street, S.W.
       Washington, D.C. 20460

         •  January 1983

-------
                                 DISCLAIMER


     This report was prepared by PEDCo Environmental, Inc. Cincinnati, Ohio,
under Contract No. 68-01-6310, Work Assignment No. 65.  It has been reviewed
by the Stationary Source Compliance Division of the Office of Air Quality
Planning and Standards, U.S. Environmental Protection Agency and approved for
publication.  Approval does not signify that the contents necessarily reflect
the views and policies of the U.S. Environmental Protection Agency.  Mention
of trade names or commercial products is not intended to constitute endorse-
ment or recommendation for use.  Copies of this report are available from the
National Technical Information Services, 5285 Port Royal Road, Springfield,
Virginia 22161.
                                      ii

-------
                                   CONTENTS
 Figures
 Tables
 Acknowledgment
 Executive Summary

 1.    Introduction

      1.1   Purpose and  Scope

           1.1.1   Level  I
           1.1.2   Level  II
           1.1.3   Level  III
           1.1.4   Level  IV

      1.2   Continuous Compliance
      1.3   Organization  of Inspection Guide
      1.4   Industry Overview
      1.5   Regulation Under the Clean Air Act

           1.5.1   State  Implementation Plan
           1.5.2   Federal standards of performance of new sources

2.   General Preparatory and Preinspection Procedures

     2.1   File Review
     2.2   Safety  Precautions

           2.2.1   Exposure to hydrogen sulfide
           2.2.2   Exposure to chlorine

     2.3  Safety and Inspection Equipment
     2.4  Preentry Observations
     2.5  On-Site Inspection Checklists

3.   Kraft Pulping Processes

     3.1  Wood Handling Department
viii
xiii
xvi
xvii

   1
   2
   2
   3
   4
   5
   7
  13

  15
  16

  21

  21
  22

  32
  32

  35
  36
  37

  53

  53
                                     111

-------
CONTENTS (continued)
                                             Page
     3.1.1  Process description

            3.1.1.1  Debarking
            3.1.1.2  Chipping
            3.1.1.3^  Knotting and screening
            3.1.1.4  Storage and transfer

     3.1.2  Sources of emissions and control
     3.1.3  Malfunctions
     3.1.4  Inspection of wood handling department

3.2  Pulping Department

     3.2.1  Process description

            3.2.1.1  Pulp digester
            3.2.1.2  Batch digester (blow)
            3.2.1.3  Digester relief and turpentine recovery
            3.2.1.4  Continuous digester
            3.2.1.5  Pulp washing
            3.2.1.6  Black liquor concentration (evaporation)
            3.2.1.7  Condensate stripping
            3.2.1.8  Black liquor oxidation

     3.2.2  Sources of emissions and control

            3.2.2.1  Digester and blow tanks
            3.2.2.2  Washer hood vents
            3.2.2.3  Evaporator condenser
            3.2.2.4  Condensate stripping
            3.2.2.5  Black liquor oxidation
            3.2.2.6  TRS scrubbers
            3.2.2.7  Incineration systems
            3.2.2.8  Lime kiln incineration

     3.2.3  Malfunctions

            3.2.3.1  Digester relief systems
            3.2.3.2  Digester blow system
            3.2.3.3  Multiple-Effect evaporators
            3.2.3.4  Black liquor oxidation
            3.2.3.5  TRS vent control system

    • 3.2.4  Inspection of pulping department

            3.2.4.1  Digester
            3.2.4.'2  Digester relief
                                               53

                                               53
                                               57
                                               60
                                               63

                                               65
                                               70
                                               70

                                               72

                                               72

                                               72
                                               74
                                               78
                                               81
                                               81
                                               84
                                               91
                                               97

                                               99

                                               99
                                              101
                                              101
                                              104
                                              104
                                              104
                                              104
                                              104

                                              107

                                              107
                                              107
                                              110
                                              111
                                              111

                                              111

                                              114
                                              115
          iv

-------
                       CONTENTS (continued)
            3.2.4.3  Brown stock washers
            3.2.4.4  Multiple-Effect evaporators
            3.2.4.5  Black liquor oxidation
            3.2.4.6  Condensate stripping
3.3  Chemical Recovery

     3.3.1  Recovery boiler

            3.3.1.1  Process description
            3.3.1.2  Sources of emissions
            3.3.1.3  Control  «
            3.3.1.4  Boiler malfunctions
            3.3.1.5  ESP malfunctions
            3.3.1.6  Inspection of recovery boiler

     3.3.2  Smelt dissolving tank

            3.3.2.1  Process description
            3.3.2.2  Sources of emission and control
            3.3.2.3  Malfunctions
            3.3.2.4  Inspection of the smelt dissolving tank area

3.4  Causticizing Department

     3.4.1  Process description
            3.4
            3.4
            3.4.1.3
            3.4.1.4
    1.1
    1.2
Green liquor preparation
White liquor preparation
Lime mud washing
Calcining
     3.4.2
     3.4.3
     3.4.4
     3.4.5
Emission sources
Control
Malfunctions
Inspection procedures
3.5  Power Boilers

     3.5.1  Process description
            3.5.1.1  Gas- and oil-fired boilers
            3.5.1.2  Coal-fired power boilers
            3.5.1.3  Wood-fired power boilers
Page

 115
 115
 122
 122

 122

 122

 126
 134
 145
 164
 167
 183

 208

 208
 210
 215
 215

 223

 223,

 225
 225
 227
 227

 228
 229
 233
 236

 243

 244

 246
 247
 249

-------
                       CONTENTS  (continued)
     3.5.2  Sources  of emissions

            3.5.2.1   Gas-fired boilers
            3.5.2.2   Oil-fired boilers
            3.5.2.3   Coal-fired boilers
            3.5.2.4   Bark boilers

     3.5.3  Control  techniques

            3.5.3.1   Gas-fired boilers
            3.5.3.2   Oil-fired boilers
            3.5.3.3   Coal-fired boilers
            3.5.3.4   Bark boilers

     3.5.4  Malfunctions

            3.5.4.1   Mechanical collectors
            3.5.4.2   Scrubbers
            3.5.4.3   Fabric filters
            3.5.4.4   Electrostatic precipitators

     3.5.5  Inspection of power boilers
            3.5.5.1
            3.5.5.2
            3.5.5.3
            3.5.5.4
            3.5.5.5

3.6  Other Sources
Opacity
Transmissometer data
Boiler operating conditions
Flue gas volume
Control equipment inspections
     3.6.1  Bleach plant

            3.6.1.1  Process description
            3.6.1.2  Sources of emission and control
            3.6.1.3  Malfunctions
            3.6.1.4  Inspection of bleach plants

     3.6.2  Raw material handling systems

            3.6.2.1  Process description
            3.6.2.2  Sources of emissions and control
            3.6.2.3  Malfunctions
            3.6.2.4  Inspection
Page

 249

 249
 249
 250
 252

 252

 252
 253
 253
 255

 255

 261
 264
 267
 270

 274

 275
 275
 276
 279
 281

 285

 285

 285
 293
 294
 294

 295

 295
 295
 297
 298
                                 VI

-------
                             CONTENTS (continued)
 4.    Compliance Determinations

      4.1   Establishing  a  Baseline

           4.1.1  Recovery boiler
           4.1.2  Smelt  tank
           4.1.3  Lime kiln
           4.1.4  Slaker
           4.1.5  Turpentine condenser and multiple-effect
                   evaporators
           4.1.6  Blow tank and hot water accumulator

      4.2   Calculation of  Emission Rates

           4.2.1  TRS sources
           4.2.2  Emissions from recovery boilers
           4.2.3  Power  boilers

      4.3   Stack Test Methods

           4.3.1   Particulate sampling
           4.3.2  TRS sampling
           4.3.3  SOp sampling
           4.3.4  Visible emissions

Appendix A     Summary of State regulations
Appendix B     EPA Reference Methods 1-5,  17
Appendix C     EPA Reference Methods 16,  16A
Appendix D     EPA Reference Method 6
Appendix E     EPA Reference Method 9
Page

 307

 309.

 312
 312
 312
 315

 315
 315

 318

 318
 318
 319

 322

 322
 323
 324
 324 '

A-l
B-l
C-l
D-l
E-l
                                    vii

-------
                                   FIGURES
Number
"                        v
 1-1      Geographic Location of Kraft Pulp Mills
 2-1      Checklist for Obtaining Information During a File Review
 2-2      Wet Scrubber Inspection Data Sheet
 2-3      Mechanical Collector Inspection Data Sheet
 2-4      Electrostatic Precipitator Inspection Data Sheet
 2-5      Fabric Filter Inspection Data Sheet
 3-1      Kraft Pulping Process
 3-2      Drum Debarking
 3-3      Bag Debarker
 3-4      Ring Barker
 3-5      Cutterhead Barker
 3-6      Hydraulic  Barker
 3-7      Knife Barker
 3-8      Norman Disk Chipper
 3-9      Drum Chipper
 3-10      Horizontal  Parallel  Chipper
 3-11      Vibratory  Chip Screen
 3-12      Gyratory Chip Screen
 3-13      Hydraulic  Dump Truck for Chips
 3-14      End-dump Freight  Car and Unloading Platform for Chips
Page
   8
  23
  39
  4.3
  45
  49
  54
  56
  56
  57
  58
  59
  59
  61
  62
  62
  64
  65
  66
  67
                                    vm

-------
                             FIGURES (continued)
Number
 3-15     Rotary Freight Car Dryer for Chips (Link Belt Limited)
 3-16     Typical Material  and Air Flow for Small  Woodyard
 3-17     Pulping Process
 3-18     Typical Digester Operating Curves
 3-19     Typical Digester Charging Room Floor
 3-20     Hot Water Accumulator Showing Primary and Secondary
            Condensers
 3-21     Typical Batch Digester Steam Flow Rate During Blow
 3-22     Worksheet for Calculation of Blow Weight (Steam)
 3-23     Digester Relief and Turpentine Recovery System
 3-24     Odor Compounds in Relief Gas after Turpentine Condenser
            as a Function of Condenser Outlet Temperature
 3-25     Continuous Digester Flow Sheet
 3-26     Vacuum Washer Flow Sheet
 3-27     Pressure Washers  Flow Sheet
 3-28     Diffusion Washer Flow Sheet
 3-29     Multiple-Effect Long-Tube Vertical Evaporators
            (Backward Feed)
 3-30     Multi-Effect Vacuum Evaporation Plant Flow Sheet
 3-31     Chart of Evaporator Temperatures
 3-32     Contaminated Condensates Air Stripping Plant Flow Sheet
 3-33     Contaminated Condensates Steam Stripping Plant Flow
            Sheet
 3-34     Stripping Effciency for Different Steam-Condensate
            Ratios with 10 Theoretical Plants
 3-35     Agitated Air Sparging System for Black Liquor Oxidation
Page
  68
  69
  73
  75
  76
  77
  78
  79
  80
  82
  83
  85
  86
  87
  89
  90
  92
  94
  95

  96-
  98
                                      ix

-------
                             FIGURES (continued)
Number                                                                  Page
 3-36     Champion Two Stage Unagitated Strong Black  Liquor
            Oxidation System                                              98
 3-37     Vaporsphere Flow Equalization Gas Holders                       102
 3-38     Floating Cover Flow Equalization Gas Holders                    102
 3-39     Hotwell  Gas Scrubber for 100 Metric  Tons Per Hour  (9400
            gpm)  Evaporation Plant for H«S-Separation of 95  Percent
            or More                     c                                 105
 3-40     Noncondensable Gas Incineration  System                          106
 3-41     Kraft Batch Digester Blow Gas Flow After Condensing and
            without Equalization                                         109
 3-42     Cross Section of B&W Recovery Boiler                           127
 3-43     Difference in Air Systems in U.S.  Recovery  Boiler  Designs       129
 3-44     Cyclone  Evaporator                                             130
 3-45     Cascade  Evaporator                                             130
 3-46     Venturi  Evaporator                                             131
 3-47     Three Types of Indirect  Contact  Evaporators                     133
 3-48     Effect of Solids Firing  Rate on  Reduced Sulfur Emissions
            and Steam Generation Efficiency                               136
 3-49     Effect of Black Liquor Solids Concentration                     137
 3-50     Effect of Black Liquor Heating Value                            137
 3-51     Bed  Temperature as a  Function of Total Air                      139
 3-52     Bed  Temperature as a  Function of Primary Air                    139
 3-53     Effect of Total  Air Supplied  to  the Unit                        140
 3-54     Theoretical  Loss of Particulate  as a Function of
            Percentage Increases in Primary Air                           141
 3-55     Effect of Primary Air Temperature                               142
 3-56     Effect of Sulfur-Sodium Ratio  in the B;lack Liquor               143
 3-57     Effect of Chlorine in Black Liquor                              144
                                     x

-------
                             FIGURES (continued)
Number                                                                   Page
 3-58     Basic Processes Involved in Electrostatic Precipitation         146
 3-59     Typical Wet-Bottom ESP with Heat Jacket                         148
 3-60     Electrical Diagram for ESP T-R Set                              149
 3-61     Typical Weighted-Wire ESP with Drag Bottom                      151
 3-62     Rigid-Frame Design                                              152
 3-63     Wet-Bottom ESP                                                  153
 3-64     Drag Chain Assembly                                             154
 3-65     MIGI Rapper Cross Section                                       156
 3-66     Internal Falling-Hammer Rapper Design                           157
 3-67     Design SCA and Efficiency of 20 Recovery Boiler ESP's           158
 3-68     Superficial Velocity Versus Year Installed                      159
 3-69     ESP Instrumentation Diagram                                     161
 3-70     Positions of Measuring Instruments                              162
 3-71     Typical Secondary Current Pattern for Unit Experiencing
            Salt Cake Buildup                                             168
 3-72     Example of a Plugged Distribution Plate                         174
 3-73    . Typical Rapper Layout on a Modern Two-Chamber Precipitator      177
 3-74     Typical Pattern Generated by Insulator Tracking                 180
 3-75     Example of Severe Corrosion of Collection Plates                181
 3-76     Example of 6-minute Average Opacity Pattern                     188
 3-77     Typical Opacity Monitor Output with Severe Rapping
            Reentrainment Losses                                          189
 3-78     Method of Calculating Additional Moisture in the Flue Gas
            Stream Due to Direct-Contact Evaporator                       196
 3-79     Optimum Secondary Current Distribution in ESP Serving Kraft
            Recovery Boiler, Assuming Uniform Rapping and Wire Size
            in All Fields                                                 200
                                     xi

-------
                   FIGURES (continued)
Number
 3-80

 3-81
 3-82
 3-83
 3-84
 3-85
 3-86
 3-87
 3-88
 3-89
 3-90
 3-91
 3-92
 3-93
 3-94
 3-95
 3-96
 3-97
 3-98
 3-99
 4-1
Secondary Current Pattern for Two ESP Chambers ; Chamber A
  is Having Maintenance Problems that Limit Power Input
Equilibrium Diagram for a Na2C03-NA2S System
Smelt Dissolving Tank with Water Sprays
                *
Smelt Dissolving Tank with Steam Shatter Jets
Wire Mesh Pad Used in Smelt Dissolving Tank Vents
Low-Energy Entrainment Scrubber for use on Smelt Dissolving
  Tank Vent (Ducon Dynamic Scrubber)
Typical Causticizing Flow Diagram
Slaker-Classifier used in Typical Causticizing Plant
H9S Emission from the Lime Kiln Related to NA?S Level in
 ^the Lime Mud
HgS Emission Related to Percent 02 in the Flue Gas
HS Emission Related to the Moisture Content in the Lime
Coal Size Distribution for Firing in a Spreader Stoker
Plugged Inlet Vane
Plugged Outlet Tube
Plugged Collecting Tube
Fabric Filter Bag Attachment Methods
Fly Ash Resistivity Curve
Flow Diagram of a Three-Stage Bleach Plant:  CEH
Cutaway of a Vacuum Washer with Short Drop Leg
Chlorine Dioxide Generating System:  Solvay
Calculation of Kraft Recovery Boiler ESP Efficiency
                                                               Page
201
209
211
212
213

214
224
226

230
230

231
258
263
265
266
269
272
289
290
292
320

-------
                                   TABLES
Number                                                                   Page
 1-1      Location of Kraft Pulp Mills in the United  States                 9
 1-2      Comparison of Light Extinction Terms                             15
 2-1      Summary of Reported Human Health Effects  of Hydrogen
            Sulfide                                                        33
 2-2      Summary of Reported Human Health Effects  of Inhalation
            of Chlorine                                                    34
 3-1      Checklist for Inspection of Wood Handling Systems                 71
 3-2      Typical Digester Liquor Requirements                             74
 3-3      Main Components of Typical  Kraft Mill  Condensates                 93
 3-4      Typical Kraft Mill  Condensate Compositions,  Mean Values
            for 10 Mills                                                   93
 3-5      Summary of TRS Control  Options for Pulping  Department             99
 3-6      Gas Flow Rates from Batch Digester                              100
 3-7      Typical Ranges of Digester Noncondensable Gas  Flow Rates         100
 3-8      Flammability Limits in Air for Kraft Noncondensable Gases        103
 *
 3-9      Flame-Spreading Velocities of Air-Mercaptan  Mixtures             103
 3-10     Typical Ranges of Evaporator Noncondensable  Gas Flow Rates       103
 3-11     Malfunctions that may Occur in Digester Relief Turpentine
            Recovery Systems                                              108
 3-12     Malfunctions that may Occur in Digester Blow Tank Hot   .
            Water Accumulator Systems                                     110
 3-13     Malfunctions that may Occur in Multiple-Effect Evaporator
            Systems                                                       112
                                    xiii

-------
                             TABLES (continued)
Number
 3-14     Malfunctions that may Occur in Black Liquor Oxidation
            Systems
 3-15     Malfunctions that may Occur in Noncondensable Gas
            Incineration System
 3-16     Checklist for Inspection of Digester Blow Systems
 3-17     Checklist for Inspection of Digester Relief Systems
 3-18     Checklist for Inspection of Brown Stock Washer Systems
 3-19     Checklist for Inspection of Multiple-Effect Evaporator
            Systems
 3-20     Checklist for Inspection of Black Liquor Oxidation
            Systems
 3-21     Checklist for Inspection of Condensate Stripping Systems
 3-22     Summary of the Effects of Key Recovery Boiler Operating
            Parameters
 3-23     Recovery Boiler Operating Parameters to be Recorded
            During Performance Tests or Inspections
 3-24     Parameters to be .Measured by the Inspector During Level III
            Inspection of Recovery Boiler ESP
 3-25     Malfunctions that may Occur in Smelt Tank Particulate
            Control Systems
 3-26     Typical Lime Kiln Mass Balance
 3-27     Power Boiler Operating Parameters to be Recorded During
            Performance Tests or Level III Inspections
 3-28     Common Letter Designations used for Bleach Agents
 3-29     Bleaching Sequences for Sulfate Pulp
 3-30     Bleaching Sequences for Hardwood Sulfate Pulp
 3-31     Inspection  Checklist for use in Bleach Plant
 3-32     Inspection  Checklist for'Material Handling Systems
 4-1      Summary of  the Effects of Recovery Boiler and  ESP Operating
            Parameters on Particulate and TRS Emission Rates
Page

 11:3

 113
 116
 118
 119

 120

 123
 125

 184

 191

 207

 215
 228

 277
 287
 288
 291
 296
 299

 313
                                      xiv

-------
Number

 4-2


 4-3


 4-4


 4-5
                             TABLES (continued)
Summary of the Effects of Smelt Tank and Venturi Scrubber
  Operating Parameters on Particulate and TRS Emission Rates

Summary of the Effects of Lime Kiln and Scrubber Operating
  Parameters on Particulate and TRS Emission Rates

Summary of the Effects of Shaker and Venturi Scrubber
  Operating Parameters on Particulate Emission Rate

Summary of the Effects of Turpentine Condenser and Multiple-
  Effect Evaporator Operating Parameters on TRS Emission
  Rate
'  Page


   314


   316


   317



   317
                                     xv

-------
                               ACKNOWLEDGMENT
     This report was prepared for the U.S. Environmental Protection Agency by
PEDCo Environmental, Inc.t Cincinnati, Ohio.  Mr. Robert Marshall was the EPA
Task Manager.  Mr. Thomas Ponder served as the Project Director, and Mr. Ronald
Hawks was the Project Manager.  The principal authors were Mr. Ronald Hawks,
Mr. Gary Saunders, Mr. Douglas Orf, and Mr. David Dunbar.
                                     xvi

-------
                               EXECUTIVE  SUMMARY
     The purpose of this inspection guide is  to  provide the  necessary  technical
information and data to support State and local  inspectors in  the  evaluation
of both new and existing kraft pulp mills with respect to continuing compliance.
The guide includes a great deal of information on the overall  operation  of
the many processes in a kraft pulp mill.  This detailed process information
is presented to provide the inspector with enough information  to allow him or
her to be conversant with plant personnel regarding the major  aspects  of a
given source or process.  By having this basic understanding of each major
process, the inspector will be able to ask informed questions  of plant per-
sonnel and to evaluate the source's program of continuous compliance.
     In addition to providing detailed process information,  the guide also
provides information on the various malfunctions that can occur within each
process and the affect that these malfunctions will have on the overall
emissions  from  the  process and in many cases  the ability of the mill  to
continuously comply with  the applicable  Federal, State, or local  regulations.
In addition to  the  discussion  of malfunctions presented  in each of the major
process subsections in  Section 3,  the reader  is  referred to the following
tables  that provide a  summary  of  the  potential malfunctions associated with
each major process  and  the effect that  these  malfunctions may have on the
operation  of the  mill  and the  emissions  from  the given process:
 Table
 3-11
 3-12

 3-13
 3-14
 3-15
 3-25
Process
Digester Relief Turpentine Recovery System
Digester Blow Tank Hot Water Accumulator
  System
Multiple-Effect Evaporator Systems
Black Liquor Oxidation Systems
Noncondensable Gas Incineration System
Smelt Tank
Page no.
   108

   110
   112
   113
   113
'   215
                                      xvn

-------
      Because the purpose and scope of State and local  agency inspections vary
 and the agency manpower is limited, the guide presents a discussion of four
 levels of compliance inspections.  Each level is increasingly more complex in
 terms of the information that should be obtained and the evaluations to be
 conducted.  Although the increasing complexity of each inspection level
 naturally increases the time required to perform each  inspection, the benefits  *
 of the more complex level  III inspections are substantial  because the cause
 of most,malfunctions and the lack of continuous compliance are not easily
 identified by Level I or II Inspections.. Briefly,  Level  I Inspections only
 consist of visual  emission evaluations.   Level  II Inspections are more detailed
 and require the inspector  to document some of the process  operating parameters
 that determine the allowable emission rate.   Basically this level  of inspection
 is commonly referred to as a "walk-through inspection" because the inspector
 walks through the  plant without physically measuring any parameters or conditions.
 He or she does, however, record any data that may be available as  a result  o.f
 plant instrumentation on the rates  of the major pieces of  process  equipment,
 the type and generating characteristics  of the  control  equipment,  major
 maintenance activities, the type and quality  of fuel-consumed,  etc.
      Level  III  inspections  require  that  the inspector  actually conduct some
 of his  or her own measurements.   Therefore, he  or she  should  have  equipment
 for measuring flue  gas  temperature,  oxygen content and  velocity, scrubber and
 fabric  filter pressure  drop,  and fan motor current and  revolutions  per minute.
 The inspector may use the data  he or she  collects to calculate gas volumes
 through the  control  devices  and  to  compare these  values with design values.
 In  general during a  Level III inspection, the inspector should record  infor-
 mation  from  plant instruments that monitor process and control device operating
 conditions including  black liquor properties, black liquor firing rates, steam
 flow, steam  pressure, furnace drafts, and such control  device parameters as
 power input  level, pressure drop, and flue gas velocity.
     Level IV Inspections are similar to Level III Inspections except that
 the former is conducted during a performance compliance stack test.  The in-
formation obtained during a Level IV Inspection is used to produce a compara-
tive baseline for future Level III Inspections so that relationship between
certain parameters and emissions can be established to  permit both inspector
and the plant personnel to monitor certain parameters to determine if the
source is complying with the applicable regulation.
                                     xviii

-------
     To assist the inspector in conducting the inspection,  the  guide  presents
a series of checklists.  The checklists for the most part are used with  Level
III Inspections.  The general information listed at the top of  the checklist
can, however, be used for a Level I Inspection if the basic information  on
the source is not already available.  The checklist can also be used  for
Level II Inspections where the plant has a considerable amount  of instrumen-
tation available for monitoring various process or control  equipment  parameters.
The reader is referred to the following checklists or data  sheets that help
to identify the type of information that should be obtained during a  Level  II
or III Inspection:
Figure              Item                                              Page
 2-1                File Review                                         23
  2-2
  2-3
  2-4
  2-5

 Table
  3-1
  3-16
  3-17
  3-18
  3-19
  3-20
  3-21
  3-23
  3-24
  3-27
  3-31
  3-32
Wet Scrubber
Mechanical Collector
Electrostatic Precipitator
Fabric Filter
Item
Wood Handling Systems
Digester  Blow Systems,
Digester  Relief Systems
Brown Stock Washer Systems
Multiple-Effect Evaporator Systems
Black Liquor Oxidation Systems
Condensate Stripper Systems
Recovery  Boiler
Recovery  Boiler Electrostatic Precipitator
Power Boiler
Bleach  Plant
Material  Handling System
  39
  43
  45
  49

Page
  71
 116
 118
 119
 120
 123
 125
 191
 207
 277
 296
 299
                                      xix

-------
      The guide also provides  (Section 4) a detailed discussion on what the
  inspector should do with all  the information obtained during the inspection.
  Procedures are presented on how to use the parameters that have been recorded
  in conducting the necessary calculations and how the field observations can
  be used to make judgements about continuous compliance.  Certain process
  parameters can be used to monitor compliance of specific sources between
  compliance tests.  The reader is referred to the following tables that provide
  information on the effects of various parameters on the overall  operation of
  the kraft pulp mill, the particulate.and TRS emissions, and the  ability of
  the source to maintain continuous compliance.
 Table               Process                                           page
  4-1
  4-2
  4-3
  4-4
  4-5
Recovery Boiler and Electrostatic Precipitator     312
Smelt Tank and Venturi Scrubber                    313
Lime Kiln and Scrubber                             315
Slaker and Venturi Scrubber                        315
Turpentine Condenser and Multiple-effect
Evaporator                                         315
      In general,  the particulate  and  S02  emission  rates from the recovery
 boiler are interrelated  because the primary method of S02 and TRS control is
 to  convert these  pollutants  to sodium sulfate, which increases the particulate
 emission rate.  The  primary  recovery  boiler parameters that affect particulate
 emissions are:  firing rate, primary  air  rate, excess air, smelt bed tempera-
 ture,  ESP power,  ESP superficial  velocity, and flue gas oxygen.  A shift in
 many of these parameters indicates an  increase in  emissions.
     The primary  parameters  that  may  be used to determine compliance from the
 smelt  dissolving  tank are connected with  the rate  of particulate generated and
 the condition of  the control devices.  Specifically the rate of generation of
 particulate is related to smelt rate  (i.e., boiler firing rate and reduction
 efficiency) and the amount of particle reentrainment.  The condition of the
 control device is related to such variables as superficial velocity and water
flow rate.
     Uncontrolled particulate emission rates from the kiln are primarily
affected by parameters that affect the superficial  velocity through the kiln,
                                     xx

-------
the particle size of the kiln dust,  rate of evolution  of  volatile  particulate,
and the feed rate to the kiln.  The  superficial  velocity  is  a  function of kiln
firing rate and temperature profile.  The rate of evolution  of volatile  par-
ticulate is related to slurry feed rate and the amount of soda present in the
slurry.  Parameters affecting the control device are liquid-to-gas ratio,
pressure drop, and particle size.  TRS emissions are related to mud washing
efficiency (% sodium sulfide expressed as Na20), flue gas oxygen,  and lime
mud slurry moisture.  The cold end temperature has an effect on TRS emission
levels.  Both the kiln excess air and temperature profile down the kiln  in-
fluence the residence time and oxidation rate of TRS compounds where the kiln
is used as a control device.
     The rate of green liquor and calcium oxide reacted in the slaker has the
                                 i
strongest effect on uncontrolled particulate emissions.  The amount of  heat
released (i.e., steam generated) and the degree of agitation are related to
the reaction rates.  The condition of, the scrubber (i.e., water flow rate,
liquor gas ratio, and pressure drop) also affects the emission rate.
     The rate of TRS emissions from multiple-effect evaporators and the
turpentine condenser is primarily a function of noncondensable gas volume
and tail gas condenser final  temperature.  The rate of TRS emissions from
the hot water accumulator  is  a function of digester operation, the condition
of the primary and  secondary  condensers, and blow gas volume.  The parameters
are generally so  interrelated that  a single parameter analysis is not effec-
tive  in predicting  emissions.  Generally,  however, TRS emissions will increase
if the condensers are plugged.
      Finally, the guide presents a  detailed discussion of establishing a
baseline that involves documenting  all  pertinent operating  parameters as they
relate to  the emission characteristics  of  the mill.   This includes both
process and  control  equipment parameters.
      The baseline may be used for several.purposes.   First, for existing
sources, baseline values may be  obtained prior  to a  stack test to  assist in
establishing representative operating  conditions.  The normal  range  of  values
may  be recorded  during  a period  prior  to a test,  and  these  values  may be
specified  in a  testing  protocol  to  establish  representative conditions  or
used  as a  starting  point in negotiating the testing  protocol  with the plant.
Comparison of documented compliance test parameters  with those specified in

                                      xxi

-------
 the protocol  helps to establish whether  the  process and control equipment
 were operating at the specified representative  conditions.  Second, for new
 sources, the  initial  compliance test establishes  the operating parameter
 values that correspond with the measured emission rate.  These values can
 then be compared with the design values.   This  provides a fixed reference
 point for comparison  to future operating data.  Third, the values of the
 baseline parameters provide data for evaluating routine inspection data.  By
 knowing the effects of the various  process and  control equipment parameters
 on emissions,  one can make comparisons to evaluate the direction and magnitude
 of any changes in performance.   Fourth,  documentation of the baseline data
 will  assist in setting specific ranges on important parameters for possible
 inclusion in an operating permit (if required by  the agency).  Finally, the
 baseline test  provides a fixed  reference  point  for comparing long-term per-
 formance trends.   Proper evaluation of the baseline data may assist in the
 establishment  of preventive maintenance  schedules as well as provide an indi-
 cation of any  design  or installation problems.  In addition, the rate at
 which the normal  operating  parameters may vary  from baseline values may assist
 the agency in  scheduling routine inspections and  periodic compliance tests.
      Baselining  should only focus on those parameters that have a documented
 affect on the  emission levels rather than  on all  possible parameters that
 might influence  the emission levels.  Collection of data that have no signifi-
 cance can be inefficient and counterproductive.   Considerable effort can be
 involved in recording  and analyzing  all process and control  equipment data
 normally available at  a  facility.  The inspector must be selective as to which
 data  to  collect.
      The use of the baseline for documenting deviations from normal  conditions
 requires the establishment  of a  logic system for each process or control
 device operating  parameter  used.  A  substantial  change in the parameter is
 evaluated  based on its  impact on the overall  emission levels.
      Once  a baseline has been established the data obtained  during a Level  III
 Inspection are usually sufficient to allow the inspector to  negotiate correc-
 tive  action with respect to the process and control  equipment without the
 expense of conducting a performance stack test.   Many deficiencies may be
 corrected as a result of increased or redirected maintenance activities.   The
ability of inspectors to negotiate such corrective action  varies  from agency

                                     xx ii

-------
to agency.  The inspector must operate within  his  agency's guidelines with
regard to negotiating compliance agreements  or issuing  notices of violation.
In some cases the corrective action can be completed  before  the  notices can
be drafted and formally issued.
                                    xxi ii

-------

-------
                                  SECTION 1
                                INTRODUCTION

     Kraft pulp mills have\been and continue to be a major source of particu-
late matter, sulfur dioxide (S02), and total reduced sulfur (TRS) compound
emissions.  Because the processes within a kraft pulp mill are numerous and
very complex, the potential emissions are a function of a number of interre-
lated process variables.  Over the past several years, a number of concerns
have been raised regarding the ability of the various sources within the kraft
pulp mill to achieve continuous compliance.

1.1  PURPOSE AND SCOPE
     This inspection guide is intended to provide the necessary technical in-
formation and data to support State and local inspectors in the evaluation of
both new and existing kraft pulp mills.  The guide includes, among other
things, a brief process description of each emission source.  This descrip-
tion is intended to provide the inspector with a basic understanding of the
technical aspects of each emission source or process.  The process descrip-
tion includes the major steps and. substeps and operating conditions that can
have substantial impact on both uncontrolled and controlled emission rates.
It also should provide the inspector with enough information to allow him or
her to be conversant with plant personnel regarding the major aspects of a
given source or process.  Having this basic understanding of each process
makes it easier for the inspector to ask informed questions of plant per-
sonnel or to seek more detailed data from the technical literature.  This
guide also should provide the inspector with the necessary information to
support certain permit stipulations or more extensive engineering evaluations.
Certain portions of the guide provide information on those process parameters
that may be used to monitor compliance of specific sources between compliance
tests.  Changes in these parameters may be used to document the need for more
extensive operation and maintenance  (O&M) to ensure continued compliance.   In

-------
  those  cases where  a  clearly  defined  cause  and effect cannot be established
  for a  given source,  certain  changes  in  various operating parameters can be
  used to support the  need for a compliance  test.  This method of source evalua-
  tion is commonly referred to as  "baselining."  The baseline is generally
  established during a period  of known compliance, typically during a compliance
  stack  test, and includes information on both the process conditions and key
  control equipment operating  parameters.
      Through an understanding of potential malfunctions, the inspector often
  can work with plant personnel to reduce both the frequency and duration of
  excess emissions and to identify and correct those malfunctions that limit
  production and cause excessive maintenance costs.
      Based on the knowledge that the purpose and scope of agency inspections
 vary and that agency manpower is limited,  four levels  of compliance inspections
 have been developed.   Each level  is increasingly more  complex in terms of the
 evaluations that must be conducted.  The increasing complexity of each level
 naturally increases the time  required to perform the inspection.   The  benefits
 of the more complex inspections  are substantial,  however,  because most mal-
 functions  and  periods of noncompliance  are  not easily  identified by Level  I
 or II  Inspections.
     The following  subsections briefly  describe  the types  of  activities asso-
 ciated  with each  level  of inspection.   More details regarding  a  given  source
 or process  are  presented in Section 3.
 1.1.1   Level I
     A Level  I  Inspection  consists  of a  visual emission evaluation according
to procedures set forth  in EPA Method 9—Visual Determination of the Opacity
of Emissions  From Stationary Sources.  The  inspector compares the opacity with
local agency  emission standards by  using appropriate averaging times and any
exclusion periods that may be set forth  in  the State or local agency's regu-
lations.
1.1.2  Level  II

     A Level II Inspection  is more detailed and requires the inspector to
document some of the process operating parameters that determine allowable
emission rates.   The additional  information includes capacity and operating

-------
rates of the major pieces of process equipment, the type and operating
characteristics of control equipment, major maintenance activities, the type
and quantity of fuel consumed, process flow diagrams, and types and quantities
of raw materials consumed.  This level of inspection is generally referred
to as a "walk-through inspection" because the inspector just walks through
the plant without physically measuring any operating parameters or conditions.
If, however, oxygen monitors, temperature charts, and etc., are available,
these data should be recorded.  Although the inspector does obtain whatever
data the plant has available on the operation of the source, which are valuable
in documenting compliance, this level of inspection may overlook serious mal-
functions or excess emission periods.
1.1.3  Level III   s
     The Level III Inspection is the most complex level of inspection that
is conducted as part of a routine inspection program.  This level is more
complete and requires substantial time, both in terms of preparation and actual
execution of the inspection.  The additional data acquired during a Level III
Inspection result in a more accurate assessment of Q&M practices.  These data
are invaluable in developing baseline information and in determining source
compliance.  Properly conducting a Level III Inspection requires that the in-
spector have equipment for measuring flue gas temperature, oxygen content,
and velocity; scrubber and fabric filter pressure drop; and fan motor current
and revolutions per minute.
     The inspector may use the data he or she obtains along with any plant
data and information to calculate gas volumes through control devices and to
compare these flue gas values with the design values.
     The inspector should also record information from plant instruments that
monitor process and control device operating conditions.  These variables may
include black liquor properties  (% solids,  heat value), black liquor firing
rates, steam flow, steam pressure, and furnace drafts and such control device
parameters as power input level, pressure drop, and  flue gas velocity.
     Plant personnel routinely record operating conditions, and  the  inspector
should compare current conditions with historic values.  Serious deviations
from established  norms should be used to evaluate emissions trends or to
determine  if a compliance test  is warranted.

-------
 1.1.4  Level IV

      A Level IV Inspection is similar to a Level III Inspection except that
 the former is conducted during a performance compliance stack test.  The in-
 formation obtained during this inspection is used to produce a comparative
 baseline for future Level III Inspections.

 1.2  CONTINUOUS COMPLIANCE
      The Clean Air Act of 1970 required all  States  to prepare a state imple-
 mentation plan (SIP) that set forth how the  State intended to attain and main-
 tain the National  Ambient Air Quality Standards (NAAQS).   Each SIP must con-
 tain among other things the necessary legal  authority and emission limitations
 to ensure attainment and maintenance of the  NAAQS.
      The Clean  Air Act of 1970 and  the initial  SIP's  developed under that Act
 placed primary  emphasis on initial  compliance of sources  with a  set of specified
 emission limitations that reduced emissions  sufficiently  to  attain the NAAQS.
 Less attention  and consideration were initially given in  the formulation of
 control  strategies that would preserve or  maintain  the air quality once attain-
 ment was  achieved.
      Control  agencies  have long recognized that initial compliance does not
 necessarily mean future or continuing compliance.   Faced  with limited  re-
 sources,  attainment dates  mandated  by legislation-,  emission  limits set by
 regulations,  and already established  source compliance dates,  the  agencies
 understandably placed  a  greater emphasis, on promoting initial  source compliance
 to ensure attainment of  NAAQS.  Now that .most sources have either  demonstrated
 initial compliance  or  are  in  the midst of  a compliance-oriented program, con-
 siderable concern  has  been raised within the air pollution control  community
with respect  to whether  a  sotfrce is operating and maintaining  its  control equip-
ment.  Some concern has  also  been raised with respect to whether a  source is
complying with the applicable emission limit on a continuous  basis.  In many
cases a source can fine  tune  its control system and make the  necessary ad-
justments to comply with an emission  limit during-a stack test conducted to
certify compliance with the applicable emission limit.  Once  these tests have
been completed, however, the control system may begin to deteriorate and the
source may no longer be in compliance with the applicable emission limit.

-------
     Reasons for the possible deterioration of the control  system include lack
of good O&M procedures, poor or virtually no maintenance, poor design,  lack of
understanding on the part of the control equipment operator, lack of reliable
instrumentation, poor recordkeeping or little or no evaluation of the records
that are kept, or lack of any desire to make the control  equipment operate
properly.
     State and local agency officials are deeply concerned with the lack of
                         ^
continuous compliance because of its potential impact on  the ability of the
State or local agency to attain and maintain the NAAQS.   As a result, many
State and local agencies are looking for ways to improve  their existing
surveillance, inspection, and enforcement programs to encourage sources; to
properly operate and maintain their control equipment; to maintain adequate
records and to use these records to avoid significant operating problems; and
to continuously comply with all applicable emission limits and visible emis-
sion standards.
1.3  ORGANIZATION OF INSPECTION GUIDE
     The inspection guide is basically divided into three major sections.
Section 2 outlines the activities that must take place before the inspector
enters the plant.  The inspector should review the files that the agency has
compiled on the types and quantities of various process equipment, type and
sizes of control equipment, results of emission stack tests, and information
from previous inspections.  These files also should contain a history of com-
plaints, malfunctions, visible emission evaluations, and previous compliance
status.  This section provides detailed checklists to assist the inspector in
obtaining the necessary information from the files.
     Section 2 also contains information concerning various safety precautions
that need to be taken during the inspection and describes the type of protec-
tive clothing the inspector should wear.  As indicated in the description of
Level III Inspections, a variety of process monitoring equipment is necessary
to obtain operating parameters for use in compliance determination.  This sec-
tion describes each piece of monitoring equipment and its purposes.
     Section 3 is by far the largest and the most useful in terms of present-
ing information that the inspector can use during the actual inspection.  This

-------
 section is divided according to the following six major processes or systems
 within a kraft pulp mill:
      o    Woodhandling
      o    Pulping
           Chemical recovery
           Causticizing
           Power boilers
o
o
o
           Other sources
      A description and process flow sheet are provided for each  of these  major
 systems.   Because kraft pulping is a complex operation,  some  of  the process
 descriptions are quite detailed.   This  detail  is  necessary to provide  the
 inspector with a sufficient understanding of the  process for  him or her to
 communicate effectively with plant personnel.
      As noted earlier, the  purpose of the compliance  inspection  is to  obtain
 information that will  allow estimation  of the  level of atmospheric emissions.
 Because many factors affect emissions,  Section 3  also includes the major
 chemical  reactions associated with each major  source  of  emissions  and  the
 techniques used to control  these  emissions.  Typical  ranges of emissions  are
 provided,  along with collection efficiencies,  flow rates,  temperatures, pressure
 drops,  power levels, etc.,  for the control equipment.
      Although systems  are designed to operate  in  a particular manner,  frequent
 malfunctions can occur and  can lead to  improper equipment  operation  and,
 utlimately,  to  control  equipment  deterioration and excess  emissions.   As  a
 result, common  malfunctions  and their causes,  symptoms,  and results  are dis-
 cussed  for each major  process  or  system within the kraft pulp mill.
      Section  3  also describes  the  procedures to be used  during the  inspection
 of specific  process equipment  and  associated control  devices.  Where appropriate,
 checklists and  tables  indicating  typical  levels of operation  are provided.
 The specific  items that need to be  observed and the specific  types of  infor-
mation  that  need  to be recorded for each  process and  control  device are also
 provided as  a further aid to the  inspector.
     Section 4  sets forth what  the  inspector should do with all of the infor-
mation  gathered  during the inspection.  This section  also outlines how to use

-------
the parameters that have been recorded in conducting the necessary calcula-
tions and how the field observations can be used to make judgment decisions.
     The table of contents for this inspection guide has been expanded to per-
mit the inspector to easily locate the necessary information regarding the
various processes within the kraft pulp mill, the control devices used to
minimize the emissions, the inspection procedures to be used for the partic-
ular process or control device, the parameters that affect emissions, and the
malfunctions that may occur.  In addition, the inspector should also consult
the rather extensive list of figures and tables that provide photographs or
sketches of process and control equipment, inspection checklists, and other
valuable information that will aid in the inspection of the source and the
subsequent evaluation of the overall source's performance with respect to
continuous compliance.
1.4  INDUSTRY OVERVIEW
     In 1979, 25 sulfite mills in the United States produced a total of
1.8 x 106 tons of pulp, 50 neutral sulfite semichemical (NSSC) mills produced
a total of 4.1 x 106 tons of pulp, and 121 kraft mills (located in 28 states)
                             fi
produced in excess of 38 x 10  tons of pulp.  Figure 1-1 shows the relative
location of the 121 kraft mills by State and Table 1-1 lists the name and
exact location of these mills.
     The type of pulping technique used depends on the end product the type
of wood available, and the general economics of the situation.  The three
major pulping techniques are described briefly.  The sulfite technique generally.
uses limestone and sulfur to produce the digestion liquor.  Liquid sulfur is
burned to form S02» which is cooled and passed through a limestone-packed
tower.  Because of its low cost, it is not necessary to recover the chemicals
from the liquor when calcium (limestone) is used.  A significant water pollu-
tion occurs, however, when the chemicals are discharged.  If sodium, ammonium,
or magnesium bases are used, however, some byproduct recovery is practical for
economic reasons.  This pulping technique is simple compared with that required
for kraft pulping.  The pulp produced by the sulfide or acid pulping technique,
which is light in color, is used for fine paper and tissue.  Further processing
                                              2
of the pulp yields rayon or cellulose acetate.

-------
00
                      NUMBER OF HILLS/STATE:

                      AlABAW . U   NUUANO . !
                      ARIZONA . 1    MICHIGAN -t
                      ARKANSAS - C   MINNESOTA - t
                      CALIFORNIA • 4  MISSISSIPPI - 4
                      FIWIM . S    MONTANA - t
                      GEORGIA . t|
                      IDAHO . 1
                      KENTUCKY - t
                      UMISIMM II
                      NUDE - 7
HEM HAMPSHIRE - I
HEM TOOK . I
MOTH CAROLINA - S
OHIO - Z
OKLAHOMA - I
OREGON - J
KHNSYIVMIA . ]
SOUTH CAROLINA - 4
TENNESSEE - 2
TEXAS - i
VIRGINIA - 4
HASHINGTCH - 7
WISCONSIN . 4
                                         Figure 1-1.    Geographic  location  of kraft  pulp  mills.

-------
        TABLE 1-1.  LOCATION OF KRAFT PULP MILLS IN THE UNITED STATES
          Company name
     Location
     Allied Paper
     James River-Dixie
       Northern, Inc.
     Champion
     Container Corp.
     Alabama Kraft
     Gulf States
     Alabama River Pulp
     Hammermill
     International Paper
     Kimberly Clark
     MacMillan Bloedel
     Scott
     Union Camp
     Southwest Forest
     Georgia Pacific
     Great Northern
     Green Bay
     International Paper
     International Paper
     Weyerhaeuser
     Crown Simpson
     Fiberboard (Louisiana Pacific)
     Louisiana Pacific
     Simpson Lee
     Alton Box
     Container Corp.
     Georgia Pacific
     Southwest Forest Industries
     Procter & Gamble
     St.  Joe Paper Company
(continued)
Jackson, Alabama

Butler, Alabama
Courtland, Alabama
Brewton, Alabama
Mahrt, Alabama
Demopolis, Alabama
Claiborne, Alabama
Selma, Alabama
Mobile, Alabama
Coosa Pines, Alabama
Pine Hill, Alabama
Mobile, Alabama
Montgomery, Alabama
Snowflake, Arizona
Crossett, Arkansas
Ashdown, Arkansas
Morrilton, Arkansas
Camden, Arkansas
Pine Bluff, Arkansas
Pine Bluff, Arkansas
Fairhaven, California
Antioch, California  -
Samoa, California
Anderson, California
Jacksonville, Florida
Fernandina Beach, Florida
Palatka, Florida
Panama City, Florida
Foley, Florida
Port St. Joe, Florida

-------
TABLE 1-1 (continued)
          Company name
     Location
     St. Regis
     St. Regis
     Continental Forest
     Continental Forest
     Brunswick
     Georgia Kraft
     Georgia Kraft
     Gilman
     Great Northern
     Interstate
     ITT Rayonier
     Owens-Illinois
     Union Camp
     Potlach
     Western Kraft
     Westvaco
     Boise Cascade
     Boise Southern
     Continental Forest
     Crown Zellerbach
     Crown Zellerbach
     Georgia Pacific
     International Paper
     International Paper
     01 in
     Pineville
     Western Kraft
     Diamond International
     Georgia Pacific
     International Paper
     Lincoln
     Scott Paper
(continued)
Jacksonville, Florida
Pensacola, Florida
Augusta* Georgia
Port Wentworth, Georgia
Brunswick, Georgia
Krannert, Georgia
Macon, Georgia
St. Mary, Georgia
Cedar Springs, Georgia
Riceboro, Georgia
Jesup, Georgia
Valdosta, Georgia
Savannah, Georgia
Lewiston, Idaho
Hawesville, Kentucky
Wickliffe, Kentucky
DeRidder, Louisiana
Elizabeth, Louisiana
Hodge, Louisiana
Bogalusa, Louisiana
St. Francisville, Louisiana
Port Hudson, Louisiana
Bastrop, Louisiana
Springhill, Louisiana
West Monroe, Louisiana
Pineville, Louisiana
Campti, Louisiana
Old Town, Maine
Woodland, Maine
Jay, Maine
Lincoln, Maine
Skowhegan, Maine
                                      10

-------
TABLE 1-1 (continued)
          Company name
        Location
     Oxford
     S.D. Warren
     Westvaco
     Mead
     Scott
     Boise Cascade

     Potlach
     International Paper
     International Paper
     International Paper
     St. Regis
     Champion         *
     Brown

     International Paper
     Champion
     Federal
     Champion

     Weyerhaeuser
     Weyerhaeuser
     Grief
     Mead
     Weyerhaeuser
     American Can
     Boise Cascade
     Crown Zeller bach
     Georgia Pacific
     International Paper
     Western Kraft
     Weyerhaeuser
   Rumford,  Maine
   Westbrook,  Maine
   Luke,  Maryland
   Escanaba, Michigan
   Muskegon, Michigan
   International Falls,
     Minnesota
   Cloquet,  Minnesota
   Moss Point, Mississippi
   Natchez,  Mississippi
   Vicksburg,  Mississippi
   Monticello, Mississippi
   Missoula, Montana
   Berlin-Gorham,  New
     Hampshire
   Ticonderoga, New  York
   Canton,  North Carolina
   Riegelwood, North Carolina
   Roanoake Rapids,  North
     Caroli na
   New Bern, North Carolina
   Plymouth, North Carolina
   Massillon,  Ohio
   Chillicothe, Ohio
   Valliant, Oklahoma
   Halsey,  Oregon
   St. Helens, Oregon
   Clatskanie, Oregon
   Toledo,  Oregon
   Gardinier,  Oregon
*  Albany,  Oregon
   Springfield, Oregon
 (continued)

-------
TABLE 1-1 (continued)
          Company name
     Location
     Appleton

     P.H. Glatfelter
     Penntech
     Bowater
     International Paper
     South Carolina
     Westvaco
     Bowater
     Packaging
     Champion
     International Paper
     Owens-Illinois
     Southland
     Southland
     Tempie-Fostex
     Chesapeake
     Continental
     Union Camp
     Westvaco
     Boise Cascade
     Crown Zellerbach
     Crown Zellerbach
     Longview
     St.  Regis
     Weyerhaeuser
     Weyerhaeuser
     Consolidated
     Great Northern
     Hammermill
     Mosinee
Roaring Springs,
  Pennsylvania
Spring Grove, Pennsylvania
Johnsonburg, Pennsylvania
Catawba, South Carolina
Georgetown, South Carolina
Florence, South Carolina
Charleston, South Carolina
Calhoun, Tennessee
Counce, Tennessee
Pasadena, Texas
Texarkana, Texas
Orange, Texas
Houston, Texas
Lufkin, Texas
Evadale, Texas
West  Point, Virginia
Hopewell, Virginia
Franklin, Virginia
Covington, Virginia
Wallula, Washington
Camas,  Washington
Port  Townsend,  Washington
Longview, Washington
Tacoma, Washington
Everett, Washington
Longview, Washington
Wisconsin  Rapids, Wisconsin
Nekoosa, Wisconsin
Kaukauna,  Wisconsin
Mosinee, Wisconsin
                                       12

-------
     The NSSC technique uses sodium sulfite and sodium carbonate or bicarbonate
as digestion chemicals.  The cooking chemicals are formed by burning sulfur to
form S02 and contacting the gas stream with sodium carbonate in an absorption
tower to yield sodium sulfite.  Because the cooking chemicals are usually not
recovered, their disposal becomes a water pollution problem.  At best, only
partial recovery of the organics is possible, despite the availability of
several techniques for this purpose unless the NSSC technique is part of a
kraft pulp mill where the NSSC liquor is often recovered in the kraft process.
The pulp, quality and pulp yield are very high.  The pulp is used primarily for
a corrugating medium because the fibers have a very high crush strength.
Hardwoods are primarily used to obtain this type of pulp.  The NSSC process
does not produce odors because of the lack of sulfide ions in the cooking
liquor.  The principal emission from this technique is SC^-
     The kraft technique, which will be explained in detail in Section 3, re-
generates the white liquor used for digestion because of the expense of the
cooking chemicals.  Although most species of wood can be pulped, the kraft
process generally uses soft woods.  The fiber produced is dark in color, but
it can be used for fine white papers after the pulp is bleached.  It is the
                                   3
most widely used pulping technique,  despite its distinctively odorous dis-
charge.

1.5  REGULATION UNDER THE CLEAN AIR ACT
     The Clean Air Act of 1970 gave the EPA the responsibility and authority
to control air pollution in the United States and its territories.  One of the
responsibilities included under Section 109 of the Act was the promulgation of
National Ambient Air Quality Standards (NAAQS).  Section-110 of the Act re-
quired the States to adopt and submit to EPA their plans for attaining and
maintaining the NAAQS in all regions of the State.  Thus, each State had to
decide which existing emission sources should be controlled and to what extent.
     In addition, Section 111 of the Clean Air Act gave EPA the authority to
develop performance standards for new stationary sources.  The New Source
Performance Standards (NSPS) established at a national level apply to both new
and modified sources.  The NSPS must reflect the degree of emission reduction
achievable through the application of the best system of continuous emission
                                      13

-------
reduction that the EPA Administrator (taking into consideration the cost of
achieving such emission reduction, and any nonair quality, health,, and en-
vironmental impact and energy requirements) determines has been adequately
demonstrated for a source category.
     Section lll(d) of the Clean Air Act also gives EPA the authority to
establish a procedure similar to that provided by Section 110 under which each
State shall submit to EPA a plan that establishes standards of performance for
any existing source for any air pollutant for which air quality criteria have
not been issued or which is not included on a list published under Section
108(a) or 112 but to which a standard of performance would apply if such
existing source were a new source.
     In addition to limiting the mass emission rate, State and local regula-
tions also limit plume opacity for certain sources.  Allowable plume opacities
vary with the jurisdictions.  Although a maximum allowable opacity of 20 percent
is common, nearly all jurisdictions allow for periodic excursions under cer-
tain conditions.  In most States the visible emission regulations are general
in wording and broad in terms of overall applicability.  Although some States
or local agencies have identified specific source categories to which a       ;
particular opacity standard would apply, the opacity requirement in most
States applies to practically all stationary sources.  The most common termi-
nology for applicability is "all existing sources."  Other terms include
"existing equipment," "existing facilities," "stationary sources," "process
operations," and "existing installations."  The use of "all existing sources"
essentially makes all sources subject to the requirement.  Although a few
States have opacity requirements that specifically apply to fuel-burning
sources or kraft pulp mills, most States have a general opacity requirement
that applies to all sources, including fuel-burning sources.
     Visible emission regulations vary widely across the United States.  The
visible emission or opacity requirements are almost evenly divided between
the Ringelmann chart and opacity.  Some State or local agencies specify only
the Ringelmann number; others specify the Ringelmann number and an equivalent
opacity limit; and still others specify only an opacity limit.   Light
transmittance (incident light flux/light flux leaving the plume), opacity,
                                         4
and Ringelmann are compared in Table 1-2.
                                      14

-------
              TABLE 1-2.   COMPARISON OF LIGHT EXTINCTION TERMS
Light transmittance, %
0
0.20
0.40
0.60
0.80
1.00
Plume opacity, %
100
80
60 .
40
-'. 20
0
Ringelmann
number
5
4
3
2
1
0
     Most State and local agencies allow visible emissions to exceed a pre-
scribed standard for some finite period of time, usually 3 to 5 minutes in
any one hour, although a few agencies allow an excursion for 6 to 8 minutes
in any one hour.  A few agencies define the excursion (or exception) as the
total minutes in a day that the standard may be exceeded.  Only a very few
agencies do not allow any excursion above the prescribed standard.
     The one exemption that applies almost universally to opacity standards
is the exclusion of uncombined water vapor from the opacity reading.  This
exemption is usually worded as follows, "where the presence of uncombined
water is the only reason for failure of an emission to meet the requirements,
                                                           A
such sections of the opacity requirements shall not apply."
1.5.1  State Implementation Plans
     Pursuant to the Clean Air Act, each State must adopt and submit to EPA a
plan that provides for attainment and maintenance of the NAAQS in all areas
of the State.  The State Implementation Plan (SIP) must include emission
limitations, schedules, timetables, and any other measures that may be necessary
to ensure attainment and maintenance of the NAAQS.  Each State determines the
mix of emissions limitations that would be applicable to the sources within
the State.
     Because a kraft pulp mill is composed of several sources that emit particu-
late, S02, and TRS, a number of regulations within a given State may be applica-
ble to some sources within a kraft pulp mill.  Table A-l in Appendix A sum-
marizes the applicable regulations for selected States where kraft pulp mills
are known to be located.
                                      15

-------
      A review of Table A-l  indicates  that  certain  sources  (such as lime kilns,
 smelt tanks,  recovery boilers,  and  power boilers)  generally are covered by
 specific regulations  that apply to  these processes, whereas other processes
 are covered under the general S02 or  particulate process regulations.  In
 addition to subjecting processes to particulate and S02 requirements, some
 States also require that  TRS to be  limited  to  so many pounds per ton of air-
 dried pulp.
 1.5.2  Federal  Standards  of Performance for New Sources
      The Clean Air Act requires that  EPA develop standards of performance for
 new stationary sources of significant air  pollution.  These standards, commonly
 known as NSPS,  are based  on the best  system of continuous emission reduction
 that has been adequately  demonstrated, taking  into account such nonair-quality
 impacts as economics  and  energy.  It  should be noted that these regulations
 take the form of standards—not just  emission  limits.  Thus, an NSPS provi-
 tion may require monitoring, process  modification, or even specific emission
 reduction methods.
      The applicable NSPS  under  Subpart BB—Standards of Performance for
 Kraft Pulp Mills—apply to  the  following affected facilities within a kraft
 pulp mill:  digester  system, brown  stock washer system, multiple-effect evapo^
 rator system,  black liquor  oxidation  system, recovery furnace, smelt dissolving
 tank,  lime kiln,  and  condensate stripper system.   In pulp mills where kraft
 pulping is combined with  neutral sulfite semichemical pulping, the provisions
 of  this subpart are applicable  when any portion of the material charged to an
 affected facility is  produced by the  kraft  pulping operation.  The NSPS for
-kraft pulp mills  indicates  the  following:
     Process
Recovery furnace
Smelt dissolving tank
Lime kiln
STANDARD OF PARTICULATE MATTER
                      Limit
       0.10 g/dscm (0.044 gr/dscf) 35% opacity
       0.1 g/kg (0.2 Ib/ton) black liquor solids
       0.15 g/dscm (0.067 gr/dscf) corrected to 10% 02
         when gaseous fuel is burned
       0.30 g/dscm (0.13 gr/dscf) corrected to 10% 02
         when liquid fossil fuel  is burned
                                       16

-------
                              STANDARD FOR TRS
Straight kraft recovery furnace


Smelt dissolving tank


Lime kiln

Digester system, brown stock washer
  system, condensate stripper system,
  or black liquor oxidation system
  that are controlled by a means other
  than combustion

Digester system, brown stock washer
  system, multiple-effect evaporator
  system, black liquor oxidation
  system, or condensate stripper
  system
5 ppm by volume on a dry basis, cor-
  rected to 8% 02
0.0084 g/kg (0.168 Ib/ton) black
  liquor solids (dry weight)

8 ppm by volume on a dry basis, cor-
  rected to 10% 02
5 ppm by volume on a dry basis, cor-
  rected to the actual 02 content of
  the untreated gas stream
5 ppm by volume on a dry basis, cor-
  rected to 10% 02*
*Unless the gases are combusted in a lime kiln subject to the NSPS provisions;
 or the gases are combusted in a recovery furnace subject to the NSPS provi-
 sions; or the gases are combusted with other waste gases in an incinerator or
 other device, or are combusted in a lime kiln or recovery furnace not subject
 to the NSPS provisions and are subjected to a minimum temperature of 1200 F
 for at least 0.5 second; or the owner or operator has demonstrated to the
 Administrator's satisfaction that incinerating the exhaust gases from a new,
 modified, or reconstructed black liquor oxidation system or brown stock
 washer system in an existing facility is technologically or economically not
 feasible.  Any exempt system will become subject to the provisions of this
 subpart if the facility is changed so that the gases can be incinerated.
     Subpart B - Adoption and Submittal of State Plans for Designated Faci-

lities indicates that after promulgation of a standard of performance for the
control of a designated pollutant from an affected facility the Administrator
shall publish a guideline document containing information pertinent to control
of the designated pollutant from designated facilities.  The guideline docu-

ment shall provide the following information:

     o    Information concerning known or suspected endangerment of public
          health or welfare caused by the designated pollutant.

     o    A description of the emission reduction systems that have, been
          adequately demonstrated.

     o    Information on the degree of emission reduction.that is achievalbe
          with each system.
                                      17

-------
      o    Incremental  periods of time normally expected to  be necessary  for
           the design,  installation,  and startup of identified control  systems.
      o    An emission  guideline that reflects  the  application of  the best
           system of emission  reduction that  has been "adequately demonstrated
           for designated facilities.
      o    Such other information that may  assist the State  in developing a
           State plan.
      Within  nine months  after notice  of availability of a final guideline docu-
ment,  each State shall adopt  and submit a  plan  for the  control of the  de-
signated pollutant  to which the guideline  document applies.   Each plan shall
include emission standards and  associated  compliance schedules.  Except as
provided by  60.24(f), the emission standards shall  be no less  stringent than
the corresponding emission guideline  specified  in  Subpart C.   Where the
Administrator  has determined  that a designated  pollutant may  cause or con-
tribute to endangerment  of public welfare  but that adverse effects on public
health have  not  been demonstrated, States may balance the emission guidelines,
compliance times, and other information against other factors  of public con-
cern in establishing emission standards.   In addition,  on a case-by-case basis
for particular designated facilities  the State may  provide for the applica-
tion of a less stringent emission standard provided that the State demonstrates
with respect to  such facility:
     o    Unreasonable cost of  control resulting from plant age, location,  or
          basic  process design.
     o    Physical impossibility of installing necessary control equipment.
     o    Other factors specific to the facility that make the application  of
          a less stringent standard significantly more reasonable.
TRS from a kraft pulp mill is a designated pollutant under lll(d)  and 40 CFR
60 Subpart B.  The recommended TRS emission limits for kraft pulp  mills
published by EPA on May 22,  1979 (44 FR 29828) are as follows:
     Digester                           5 ppm
     Multiple-Effect Evaporators        5 ppm
     Straight Recovery Furnace          5 ppm
       System Designed  for Low TRS
       Emissions
                                     18

-------
All other Straight Kraft Recovery
  Furnace Systems
Cross Recovery Furnace Systems
Lime Kiln Systems
Condenser Stripper System
Smelt Dissolving System
20 ppm

25 ppm
20 ppm
 5 ppm
0.084 g/kg of black liquor solids
  (dry weight)
                                  19

-------
                          REFERENCES FOR SECTION 1
1.   U.S. Environmental Protection Agency.  Standards Support and Environ-
     mental Impact Statement, Vol. I:  Proposed Standards of Performance for
     Kraft Pulp Mills (Revised).  EPA-450/2-76-014a, September 1976.

2.   U.S. Environmental Protection Agency.  Technology Transfer.  Environ-
     mental Pollution Control, Pulp and Paper Industry, Part I, Air.,  EPA-
     625/7-76-001, October 1976.

3.   Kirk-Othmer.  Encyclopedia of Chemical Terminology, 2nd Ed.  Vol. 16,
     1970.

4.   U.S. Environmental Protection Agency.  Analysis of State and Federal
     Particulate and Visible Emission Combustion Sources.  EPA-450/2-81-080,
     November 1981.
                                      20

-------
                                  SECTION 2
              GENERAL PREPARATORY AND PREINSPECTION PROCEDURES

     Preparation is the key to a successful inspection of a kraft pulp mill.
Such preparation includes:
     o    Becoming familiar with the various types of process and control
          equipment used
     o    Reviewing past operating practices
     o    Procuring and testing the necessary inspection equipment to be sure
          it is working properly
     o    Inspecting the plant's exterior to obtain information about oper-
          ating practices
     o    Advising key plant personnel well in advance so that they are avail-
          able to answer questions and take part in the inspection.  The
          cooperation of key plant personnel is critical to the success of
          the inspection.
Advanced preparation on the part of the inspector can save valuable time for
both the inspector and plant personnel.  A well-informed and prepared inspec-
tor generates a degree of confidence that makes plant personnel more inclined
to provide information critical to completing a comprehensive plant inspec-
tion.

2.1  FILE REVIEW
     Baseline operating parameters for both process and control equipment
should be obtained from information filed at the agency prior to the inspec-
tion.  The typical source file should contain information on permit
activity, previous inspections, or emission stack tests.  The files generally
provide descriptions of such characteristics as the size, throughput, and
efficiency of the process and control equipment, which give the inspector
a perspective on the overall layout and operation of the mill.  The source
files also contain information on citizen complaints, equipment malfunctions,
                                     21

-------
 opacity levels, and the overall  compliance status of the various  sources  at
 the mill.   This information helps to establish a history of the processes and
 helps the inspector to focus attention on those processes with continuing com-
 pliance problems.
      File data on  previous emission stack tests and other inspections  should
 be used to establish a baseline  for comparison with future inspections  or
 stack tests.   Such information also helps to  establish  a normal operating    '
 range for both process and control  equipment  and permits the  inspector  to
 readily note any deviations.
      The sample checklist in Figure 2-1 should be used  to gather  information
 from the agency files about previous abatement activities.

 2.2  SAFETY PRECAUTIONS                                                      ;
      Safety precautions must be  practiced during plant  inspections because
 heavy equipment movement,  high-temperature process  equipment, high-pressure
 steam,  toxic gases,  and noise are common.   Because  kraft pulp mills have  a
 significant number of high-temperature  processes and many of the  processes
 are wet, extreme caution  should  be  taken  to avoid burns  from high-temperature
 processes  and  the  possibility of slipping  and  falling around the  wet processes.
      There are also  several  specific processes  that are  of concern in terms
 of  the  inspector's overall  safety.   The operation of debarkers and chippers
 present safety hazards  because of the potential  for small  bark and wood
 particles  to be ejected from  these  processes at  extremely  high rates of speed.
 In  addition, these processes  are extremely noisy, which makes it  difficult to
 hear any vehicles  that may  be moving  through the  mill.
     Another special  concern  is  the  potential for an explosion of the smelt
 dissolving  tank or recovery boiler.  When molten  smelt at  a temperature of
 approximately  1600°F  is added to smelt dissolving tanks containing water and
green liquor,  the instantaneous transfer of heat between  the smelt and the
 liquid produces an explosion of steam.  These explosions can be on the surface
of the tank or deep within the tank.  Surface explosions result in a shower
of smelt and green liquor that can be hazardous because of the high tempera-
ture of the material.  Explosions deep within the tank,  however,  may blow the
tank lid off or split the tank at the seam.  Both types of explosion can result
                                     22

-------
NAME OF COMPANY

ADDRESS
PLANT CONTACT
PREVIOUS INSPECTIONS:

  Date            Process(es)
                                     Comment(s)
  Date
Process(es)
Comment(s)
          Figure 2-1.  Checklist for obtaining information during
                        a file review.  (Continued)
                                    23

-------
 Date
 Process(es)
                                                     Comment(s)
Date
Process(es)
                                                     Comment(s)
        Figure 2-1.  Checklist for obtaining information during
                      a file review.  (Continued)
                                  24

-------
STACK TEST:  (attach specific test results)
                                                           Emission  rate
  Date
Process(es)
Process rate
Actual
                                                                     Allowed
COMPLAINTS:

  Date
        Source
                       Comment
MALFUNCTIONS:
   Date
        Source
                                                          Comment
            Figure 2-1.   Checklist for obtaining  information  during
                          a file review.   (Continued)
                                      25

-------
PROCESS  INFORMATION:
B.
     RAW MATERIAL RECEIVING (Wood yard)
     	Round wood
     	 Chips
     	 Sawdust
DEBARKING
Type	
                     Number
     CHIPPER
     Type
                Number
     CHIP  UNLOADING

    	Truck
    CHIP TRANSFER

    	 Conveyor
    DIGESTERS
    Type	
              batch
              Continuous
            Type
Process rate
                                                             _tons/h
Process rate
                                                             _tons/h
                            Rail
          Barge
                              Pneumatic
                              Receiver 	
                              Control type
                              Process rate
                                                           tons/h
Number
Cook time
Pulp rate
Volume
min
ton/h
                                                                  ft*
                          Cooks  per day 	
                          Number of blow  tanks
                          Volume	 ft"
                          Condensers primary __
                                   secondary 	
                          TRS control 	
                          Number __	
                          Pulp rate ______
                                                     yes
                             no
                         TRS control
                       Scrubber
                                                   tons/h
                                                     yes
                             no
         Incinerator
         Figure 2-1.   Checklist for obtaining information  during
                       a file review.   (Continued)
                                    26

-------
     BLACK LIQUOR OXIDATION
Weak
Air
Type
Strong
Oo

     Inlet sulfidity
     Efficiency 	
     TRS control 	
          Type	
     g/liter    Outlet sulfidity
                           g/liter
  yes
no
Scrubber
   Incinerator
H.    MULTIPLE EFFECT EVAPORATOR
     Number of lines 	
     Number of effects 	
     Inlet black liquor solids (BLS)
     Feed rate	M Ib/h
     TRS control	yes  	
                       Outlet BLS
              no
     RECOVERY BOILER (complete for each boiler)
     Boiler manufacturer	
     Design	old  	
     Evaporator type _,	
         new
      Cascade
      Cyclone
      Venturi
      Noncontact
     Inlet BLS to evaporator
     BLS to guns	3
     Firing rate BLS 	
                 BL  	
     Rated capacity  	
     Liquor heat value _
     Auxiliary fields	
     Particulate control
      M Ib/h
      gpm
      Ib steam/h
      BLS/h
      BLS/day
      tons air-dried pulp (TADP)/day
     _ Btu/lb BLS
     _ Natural gas	Residual oil
       .-  ESP
          Scrubber
           Figure 2-1.  Checklist for obtaining information during
                         a file review.   (Continued)
                                      27

-------
 0.    SMELT DISSOLVING TANK (complete for each tank)
      Number of tanks
      Smelt rate 	
      Control
        tons/h
        Mesh pads         Scrubbing media
        Low energy scrubber
        Venturi scrubber
        None
      LIME  KILN  (complete  for each  kiln)
      Capacity
      Fuel
      Heat  input 	
      Product rate 	
      Slurry solids 	
      Slurry feed rate
      Slurry soda 	
     tons/day
       10° Btu/h
       _ tons/day
             gpm
     Particulate control  	
     TRS control 	yes
               Venturi
                    no
     Low energy scrubber
     SLAKER  (complete for each slaker)
     Capacity	tons/h
     Green liquor 	
     CaO
     Control
     Showers
Low energy scrubber
M.   POWER BOILERS (complete for each boiler)
     Type	
     Heat input
     Fuel
     Control
	 10° Btu/h
 Coal  	Wood
	 Multicyclone
	 Venturi
	ESP
	 Baghouse
     Oil
Gas
           Figure 2-1.   Checklist for obtaining information during
                         a file review.   (Continued)
                                     28

-------
N.   BLEACH PLANT

     Capacity 	
TADP/day
     Number of stages
     Bleach chemicals
     Control
        C12
        cio2
        Other
 Packed bed
 Other
           Figure 2-1.  Checklist for obtaining information during
                         a file review.  (Continued)
                                      29

-------
  in death or serious  injury to those  in the area.  If green liquor represents a
  high percentage of the liquid, the explosion is more violent and can occur at
  a higher temperature because the salts in the green liquor lower the boiling
  point of the liquid.  The higher the temperature of the liquid, the less likely
  and less violent will be the explosion.1'2
      Recovery boiler explosions may occur for many reasons.   The most common
  is water/smelt interaction as a result of water tube leaks,  firing of weak
  liquor, or the improper use of.water or air lances.   Between 1969 and 1973
  there were 23 recovery boiler explosions caused by smelt-water contact and
  three caused by auxiliary fuel  firing practices.
      It should be noted that the possibility of a recovery boiler explosion
 does exist and the inspector should be familiar with the specified evacuation
 routes and procedures before entering the recovery boiler area.2
      Turpentine vapors  and other organic mists  emitted  from  the digesters  also
 create an explosion  hazard.   Scrubbers can be used to reduce  this explosion
 potential.
      A major safety  concern  at  kraft  pulp mills  is the  entry  into a  confined
 area.   The  cardinal  rule  for  entering a  confined  area is "never trust your
 senses."  What  may appear to  be  a harmless situation may well  be a potential
 threat.   The  three most common conditions constituting  a threat are:
      o    Oxygen deficiency                          '                   .      i
      o    Presence of combustible gases and vapors                            ;
      o    Presence of toxic gases and  vapors.
     An inspector  should  always anticipate that any one  or a combination of
 the above conditions might exist in a confined area such as ductwork, stack,
 open tanks, penthouse, or the internal portion of a wet  scrubber or an ESP.
 Tests for flammability, oxygen deficiency, and toxicity must be made before
 an inspector enters a confined area.  No  one factor will provide more safety
 than the knowledge of the potential  threats that may exist within the area
 to be inspected.  Armed with this knowledge, the inspector can take appropriate
 precautions and use the proper equipment  to minimize any potential dangers.
     Many liquids used and handled in the mill are corrosive  or a skin irri-
 tant.   Care must be taken to avoid contact with such chemical  reagents or
bleach chemicals as:   sodium hydroxide (NaOH), sulfuric  acid  (HgSOj, black
                                     30

-------
liquor, and white liquor.  Many solids handled at the mill  are also hazardous
to the skin and eyes and contact should be avoided.   The inspector should
wear appropriate eye protection (glasses and side shields)  when conducting
inspections in the mill.
     Because pulping is carried out at elevated temperatures and steam is used,
accidental contact with heated surfaces may occur.  The inspector should be
constantly aware of the location of piping, duct work, or equipment that may
present a potential hazard.  Protective clothing (long sleeved shirts, gloves,
etc.) should be worn to minimize burns as a result of accidental contacts with
hot surfaces.
     The inspector should remove jewlery, ties or other loose objects before
entering the work area.  Because of the close quarters in many mills and the
possible contact with moving equipment during the process of taking measure-
ments, any object that  could become easily entangled should be removed.
     In addition to the general concerns noted above, if an inspector is con-
ducting an internal inspection of a wet scrubber or an ESP, he/she should:
     o    Observe interlock procedures
     o    Observe the confined entry  procedures of isolation lockout
     o    Watch footing
     o    Never work alone
     o    Wear protective  equipment
     o    Deenergize unit  before  entry
     o     Purge  unit before entry
     o    Use grounding straps
     o    Never  enter  full or partially full  hoppers (wet- or dry-bottom).
      Lastly, the inspector should be aware of and obey all safety requirements
 set forth by plant'personnel.   Many mills have their own safety procedures
 that must be obeyed;  therefore,  the inspector should meet  briefly with plant
 personnel regarding any additional  safety concerns or requirements that the
 mill has established with respect to access to certain equipment or areas of
 the plant and any special  safety equipment that may be required.
                                      31

-------
 2.2.1  Exposure to Hydrogen Sulfide
      Because of the number of sources in the mill  that produce TRS emissions
 and the potential  for the inspector to come in contact with these  gases,  the  ;
 following is provided to inform the inspector of the effects of hydrogen
 sulfide exposure.   Hydrogen sulfide is a colorless gas with an obnoxious  odor
 at low concentrations.   In humans,  it can cause headache,  conjunctivitis,
 sleeplessness,  pain in  the eyes at  low concentrations, and death at high  con-
 centrations.  Hydrogen  sulfide is extremely toxic  to humans.   Initial  exposure
 is through the  respiratory tract, from which the hydrogen  sulfide  is carried
 by the blood stream to  various body organs.   Hydrogen sulfide in the blood
 can block oxygen transfer at high concentrations.-   Prolonged exposure can
 also result in  enzyme poisoning and irreversible nerve tissue damage.  '   At
 high concentrations (>1,000,000 yg/m )  exposure causes death by paralysis of
 the respiratory center.    Table 2-1 summarizes  the health  effects  associated
 with human exposure to  hydrogen sulfide.
      The  reported  odor  threshold for hydrogen  sulfide varies  from  1 to 45
 yg/m .    At 500 yg/m ,  the odor is  distinct; at 4,000 to 8,000 yg/m3,  it
is offensive and  intense.  At  30,000 to  50,000 yg/m  , the odor is very strong
              fi                             *3
but tolerable;  at greater than  320,000  yg/m , the smell is less pungent
because the olfactory nerves become paralyzed.  Continued exposure to high
concentrations results in a distinct reduction of odor perception as a result
                     3                                        ?
of olfactory fatigue.   At concentrations above 1,120,000 yg/m, no odor may
be sensed and death can occur  rapidly.   Dulling of  the olfactory nerves
constitutes a major danger to  the inspector who is exposed to moderate to high
concentrations for extended periods of time.
2.2.2  Exposure to Chlorine
     The inspector should be particularly cautious while inspecting the bleach
plant, where chlorine and chlorine dioxide may be used as bleaching agents.
The extraction and washing of  bleached pulp exposes  the pulp to ambient,air,
where residual chlorine gas may be lost.  Although these sources are hooded
and vented, the possibility of exposure  still exists in some cases.
     Chlorine, a dense, greenish-yellow  gas with a distinctive irritating
odor, is a very strong oxidizing agent.  This highly corrosive gas is
extremely hazardous to all life forms.

                                      32

-------
           TABLE 2-1.  SUMMARY OF REPORTED HUMAN HEALTH EFFECTS OF
                              HYDROGEN SULFIDE7
   Concentration,
       yg/m3
                    Effect
        1  -  45
          10
         150
         500
      15,000
      30,000

   30,000 -  60,000

     150,000

  270,000 -  480,000

  640,000 -  1,120,000

1,160,000 -  1,370,000

    > 1,500,000
Odor threshold.
Threshold or reflex effect on eye.
Smell slightly perceptible.
Smell definitely perceptible.
Minimum concentration causing eye irritation.
Maximum allowable occupational exposure for
  8 hours.
Strongly perceptible but not-intolerable
  smell.  Minimum concentration causing lung
  irritation.
Olfactory fatigue in 2 to  15 minutes; irrita-
  tion of eyes and respiratory tract after
  1 hour; death in 8 to 48 hours.
No serious damage for 1 hour but  intense local
  irritation;  eye irritation in 6 to 8
  minutes.
Dangerous concentration after 30 minutes or
 . less.
Rapid unconsciousness, respiratory arrest, and
  death, possibly without  odor sensation.
Immediate unconsciousness  and rapid death.
     Human sensitivity to chlorine gas varies greatly.  Its main effect is an
irritating and corrosive attack on the mucus membranes of the eyes, nose,
throat, and respiratory tract.10  High concentrations of chlorine can damage
the lungs and result in pneumonia,11 emphysema,   and bronchitis.  In very
                                                                      12 13
high concentrations, damage may be severe enough to cause suffocation.  '
Table 2-2 summarizes the health effects attributed to chlorine exposure.
                                      33

-------
            TABLE 2-2.  SUMMARY OF REPORTED HUMAN HEALTH EFFECTS OF
                            INHALATION OF CHLORINE9
Concentration,
     PPM
 Exposure time
                Effect
    < 1.0
      1

     3.3

     3-5

     3-6
     4

     5
     5
     5

    10
  > 10
   14-21
    20
    50
   100
 Several hours
Working
   conditions
30-60 minutes


< 1 minute

0.5-1.0 hour
< 30 minutes
30-60 minutes
< 1 minute
 Objective symptoms of irritation.
 Slight; symptoms after several  hours
   exposure.
 Risk to health or life;  impossible
   working conditions.
 Tolerable for short periods  of time
   without objective evidence or
   injury.
 Stinging or  burning sensation  in the
   eyes, nose,  and throat;  sometimes
   headache due to irritation of the
   nasal sinuses.
 Maximum amount that can  be inhaled
   for 1 hour without serious
   disturbances.
 Slight smarting of the eyes  and irra-
   tation of  the nose and throat.
 Premature  aging;  those exposed  suffer
   from disease of bronchi  and become
   predisposed  to  tuberculosis;  teeth
   corrode  from hydrochloric  acid;
   inflammation or ulceration of the
   mucous membrane of the nose occurs.
 Does  not endanger life.
 Noxious  effect; impossible to breathe
   after  several minutes.
 Severe coughing and  eye  irritation.
 Immediate and  delayed.
 Dangerous.
 Endangers life.
 Immediately fatal.
Cannot be tolerated for longer than
  1 minute.
                                    34

-------
                                                                      o
     The odor threshold reported in the literature is below 1,000 yg/m  (0.33
ppm), but it may be lower when acting in combination with hydrogen chloride.  '
The 8-hour threshold limit value (TLV) established by the National Institute
of Occupational Safety and Health  (NIOSH) is 3 mg/m3 (1 ppm).16  A peak of
0.5 ppm/15 min has been recommended but not adopted.  The inspector should
wear respiratory protection while  in the bleach plant area and when working
near process equipment and control devices.  The plant will usually supply
respirators for visitors and inspectors to use.

2.3  SAFETY AND INSPECTION EQUIPMENT
     While conducting an inspection of the pulp mill, the inspector should
use the appropriate protective clothing and safety equipment and  follow all
company rules and recommendations.  Because of the potential for  flying pieces
of wood and bark particles, the  inspector should wear safety glasses with
sideshields for protection.  The debarkers, chippers, and various conveying
systems are the major sources of these flying wood or bark particles.  Hear-
ing protection such as ear plugs should also be used because of the high
noise  levels around debarkers and  chippers.  Steel-toed shoes and a hard hat   .
are required for protection against overhead hazards and heavy objects.  A
nonslip sole shoe with ankle support  is'recommended.  A long-sleeved shirt,
gloves, and trousers should be worn for protection  from the high-temperature
processes  and  the rough  texture  of the delivered raw materials  (wood).  Neck-
ties,  hair ribbons, rings, etc., should be removed  prior to the actual inspec-
tion.  Dust and mist respirators should be available and used around potentially
dusty  operations.   In some cases a gas mask may be  required.   In  a  few cases,
the  inspector  should use a self-contained  breathing apparatus when  required
^to  enter  a confined area.
     The  equipment  or  instruments  used  during  an  inspection vary  according  to
the  time  allotted and  the  level  of the  inspection.   For  example,  a  detailed
Level  III  Inspection  involving  several  days  at the  mill  requires  the  following:
a pitot  tube  and  manometer for  measuring  the  gas  stream  flow  from each pro-
 cess,  a manometer for  measuring the  pressure  drop  across  the  appropriate
 control  equipment,  a  thermometer or thermocouple  for measuring  stack  gas
 temperatures,  a wet bulb/dry bulb thermometer and  psychrometric chart for
 determining  moisture,  a tachometer for measuring  fan  speed,  an  ammeter for

                                       35

-------
 measuring fan motor current,  an oxygen meter for determining  concentration  of
 the exhaust gases,  and a Fyrite or Orsat for determining  gas  composition.
 Some general  pieces of equipment are also necessary,  including  1) a  flash
 light,  2) a stopwatch, 3) a  tape measure, 4) a  pressure gauge,  and 5)  a     ;
 water-flow device.                                                          ;
      For a less  detailed Level  II Inspection, the inspector may only need a
 camera,  a compass,  and a stopwatch.   During  a Level  II Inspection, when  the
 inspector does not  make any measurements, he or she  should obtain readings
 from plant instruments (e.g., pressure in the digesters,  fuel firing rates,
 temperatures  of  stack  gases)  and review the  data.
      If  permitted on mill  property,  a camera can  be  useful to document ex-  i
 cessive  opacity  levels and to provide graphical  descriptions of problems
 arising  from  poor maintenance and housekeeping, missing bags, or control
 equipment components,  and the relative location of certain sources.  Immed-
 iately after  taking  a  photograph,  the inspector should make a log book
 notation  describing  the situation represented in  each photograph and the
 time, date, weather  conditions,  and  pertinent directional information.
     A compass is useful  for determining  directions of sources  relative to
 each other, to the sun,  and to  the inspector.  A  stopwatch for  timing  visible
 emissions  observations is-also  useful.                                      •

 2.4  PREENTRY OBSERVATIONS
     Before entering the  mill property or while moving from one process opera-
 tion to another, the inspector  can gain considerable information by  preentry
 observation of the mill.   Sources  of  fugitive dust can sometimes best  be
 observed outside the mill.  The  inspector should also note the weather con-
 ditions  (especially precipitation  and  windspeed) during and, prior to the
 inspection.
     Exterior observations also  provide an opportunity to observe the  general
 housekeeping practices  of  the mill and give  the inspector an overall  picture
of the mill layout for  comparison with  information obtained from the files.
The inspector also.can  get an idea of  the level  of activity by observing the
raw material and pulp moving operations, mill traffic, and processes.
                                      36

-------
     While outside the mill property, the inspector can normally use a camera
to photograph excessive visible emissions.  Some state laws, however, prohibit
the use of a camera witho'ut the prior permission of the mill.  Data regarding
any photographs that may be taken (e.g., date, time of day, weather conditions,
position relative to the source) must be recorded immediately.  In all cases,
while on plant property, the inspector must get permission from plant personnel
before taking photographs.
     Visible emission observations are important in determining the operating
conditions of some processes and their associated control equipment.  When
observing opacity from stacks, the inspector should follow the procedures of
Federal EPA Method 9 or appropriate State agency method.  Windspeed, sky con-
dition, and other weather  data should be recorded for future use because the
reading may be challenged  in court.  A diagram is also important in  identify-
ing the particular source  being observed  (e.g., the No.  3 coal-fired power
boiler) and the observer's position in relation to the sun and the  source.
     The  inspector should  record opacity readings on the observation form for
a  specified duration, depending on the local requirements.   Although the
regulation may specify a  plume opacity below a certain average for  a 6-minute
period, the inspector may  want to take the  reading for a longer period, say
30 minutes, and determine  if a 6-minute period may have  exceeded  the applica-
ble  limit.
     Opacity  readings are sometimes  best  obtained  before the inspector enters
the  mill  for  the  inspection or'after  he/she leaves the  plant property.  The
inspector should  compare  the  visual  measurements with  values obtained by  using
the  plant's continuous  emission  monitoring  equipment  (if available) for the
same time period.   The  frequency of  calibration  of these continuous emission
monitors  should  be noted.
      If the agency's policy  is  to provide the mill with a  copy of the opacity
readings  taken during the inspection, the plant official receiving the copy
should sign  and  date the  original  of the opacity readings.

2.5  ON-SITE INSPECTION CHECKLISTS
      During the on-site inspection,  the inspector may find it useful to have a
 series of checklists to record the information obtained during the inspection.
                                      37

-------
Figures 2-2 through 2-5 are examples of the types of check lists that can be
used to obtain information on the process and abatement equipment;   Addi-
tional copies of certain sections of the check lists should be completed if
there are multiple sources for each process (i.e., three recovery boilers,
three lime kilns, etc.).
                                      38

-------
LOCATION	
DESIGNATION	
INSPECTOR(S)        :
CLAIMED CONFIDENTIAL   Yes
           DATA SHEET NO.
           DATE
No
A.   DESCRIPTIVE INFORMATION
     Wet Scrubber Type	
     Manufacturer 	
     Model Number 	
     Date Installed	
     Process/Source Controlled _
     Particulate Characteristics
 B.   COMPONENT  INFORMATION  (Describe if applicable)
     1.   Gas Pretreatment:
           Presaturator
           Cyclones 	
           Settling  Chamber
           Other      	
           Demister:
           Cyclone 	
           Chevron 	
           Fiberous  Mat
           Other
           Pumps:
           Number 	
           Recirculation 	
            Pump Manufacturer
           Recirculation
            Pump Rated Horsepower
           Recirculation Pump Type
          Figure 2-2.   Wet scrubber  inspection data sheet.   (Continued)
                                       39

-------
                                                  Data Sheet No.
                                                  Preparer
                                                  Confidential:"
                      Yes
         No
B.   COMPONENT INFORMATION (continued)

     4.   Fan/Motor (Specify)

          Fan Manufacturer 	

          Blade Type:   Radial _____

          Drive:   Direct
Backward
Forward
          Damper Position 	

          Motor Manufacturer
          Model  No.
         Belt
          Rated Horsepower 	

          Location:   Forced  Draft
         Induced Draft
     5.    Instrumentation  (Check  if  Applicable)

          Differential
          Pressures:
         Temperatures;
         pH:
         Flow Rates;
         Motor Current:
mroat
Separator
Demister
Gas Outlet
Gas Inlet
Liquor Inlet
Liquor Outlet
in.
in.
in.
°F
°F
°F
°F
Recirculation
Exit Liquor
Fan Motor Current
Other
Nozzle Pressure
Recirculation
Makeup
Purge
Fan
Pump
qpm
gpm
anm
a
a
                                                            H20

                                                            H20

                                                            H20
       Figure 2-2.  Wet scrubber inspection data sheet.   (Continued)
                                    40

-------
                                                  Data Sheet No.
                                                  Preparer
                                                  Confidential:   Yes
No
B.   COMPONENT INFORMATION (continued)
     6.   Materials of Construction (Specify type and gauge)
          Presaturator	
          Throat	\	
          Scrubber Shell
          Trays/Bed Supports
          Mist Eliminator 	
          Fan Housing 	
        Figure  2-2.  Wet scrubber  inspection  data  sheet.   (Continued)

                                     41

-------
                                                  Data Sheet No.
                                                  Preparer
                                                  Confi dential:   Yes
No
C.   DIAGRAM
     1.  Sketch wet scrubber system.   (Show all  major components  and  processes
         controlled.)
     2.   Sketch wet scrubber layout
        Figure 2-2.  Wet scrubber inspection data sheet.   (Continued)

                                     42

-------
LOCATION
DESIGNATION
DATE
                                   DATA SHEET NO.
                                   INSPECTOR(S)
CLAIMED
CONFIDENTIAL
    Yes
No
A.   DESCRIPTIVE INFORMATION
     Mechanical Collector Type
          Cyclone 	
          Cyclone Bank
          Multiclone
                         Settling Chamber 	
                         Double Vortex Cyclone
                         Other (describe) 	
     Manufacturer
     Model Number
     Date Installed
     Process/Source Controlled
     Particulate Characteristics
B.   COMPONENT INFORMATION
     1,
Cyclone
Diameter of Body
Cone Length 	
                                                                   ft
                                                                   ft
          Material of Construction
          Gauge of Metal 	
          Number of Cyclones
          Hoppers
          Number	
          Slope 	
          Insulation:  Yes
          Heated:  Yes 	
          Vibrators:
            Yes
                      No
                      No
                      No
    Figure 2-3.  Mechanical collector inspection data sheet.   (Continued)
                                      43

-------
                                                   Data Sheet No.
                                                   Preparer
 B.
                                              Confidential:Yes _
COMPONENT INFORMATION (continued)
 3. Solids Removal (Check applicable items and provide dimensions)
                                                                          No
     4.
           Rotary Valves 	
           Flapper Valves 	
           Screw  Conveyors 	
           Pneumatic  Conveyors
           Free Fall
      Fan/Motor
      Fan Manufacturer
      Model  Number 	
      Blade  Type:
      Drive:
                       Radial
                  Direct
Backward 	
       Belt
Forward
          Motor Manufacturer
          Model Number
          Rated Horsepower
          RPM
          Location:  Forced Draft
                                          Induced Draft
C.   SYSTEM LAYOUT
    Figure  2-3.  Mechanical collector  inspection data  sheet.   (Continued)

                                    44

-------
LOCATION 	
DESIGNATION 	
INSPECTOR(S) 	
CLAIMED
CONFIDENTIAL Yes
                     DATE SHEET NO.
                     DATE
No
A.   DESCRIPTIVE INFORMATION
     Electrostatic Precipitator Type
     Manufacturer 	
     Model No.
     Date Installed
     No. of TR Sets
     No. of Fields in Series
     Plate Length, Height 	
     Plate Area
     Plate Spacing
     Number of chambers
Bypass System:  Yes
                                                    ft
                                                    ft2
                                                    in.
      No
B.   ELECTROSTATIC PRECIPITATOR LAYOUT SKETCH (each square 2 ft x 2 ft)
  Figure 2-4.  Electrostatic precipitator inspection data sheet.  (Continued)
                                      45

-------
C.   COMPONENT INFORMATION
     1.
     2.
Gas Distribution
Perforated Plates:  Inlet
Turning Vanes:  Inlet 	
Rappers
Discharge Wires
     Type	
               Number
               Manufacturer	
               Air Pressure (if applicable)
          Collection Plates
               Type	:	
               Number
               Manufacturer
                                                  Data  Sheet No. 	
                                                  Preparer
                                                  Confidential:Yes
                                                               No
                                                      Outlet
                                                      Outlet
               Air Pressure  (if applicable)
           Distribution  Plates/Vanes
               Type	,	
                Number
                Manufacturer
                Air Pressure  (if  applicable)
  Figure 2-4.  Electrostatic precipitator inspection  data  sheet.   (Continued)

                                       46

-------
                                                  Data Sheet No.
                                                  Preparer
                                                  Conf i dentTaT:   Yes
                                                          No
C.   COMPONENT INFORMATION (continued)
     Layout of Rappers (Show discharge wire rappers as D, collection plate
     rappers as C, and distribution plate rappers as X.)
          3.
Hoppers
Number _
Length _
Slope
               Level  Indicator  Height
               Insulation:   Yes	
               Heating:   Yes	
ft
degrees
ft  from top
                            No
                            No
Figure 2-4.  Electrostatic precipitator inspection data sheet.  (Continued)
                                     47

-------
c.
                                                  Data Sheet No.
                                                  Preparer
                                                  Confidential:Yes
                                                                    No
COMPONENT INFORMATION (continued)
4.   Dry Bottom (Recovery Boiler)
     Drag Chains:  Longitude 	
                   Latitude
     6.
     7.
     8.
                        Number of Chains
          Wet Bottom (Recovery Boiler)
          Number of Agitators 	
 Drag  Chain  Motor  Current	
 Drag  Chain  Speed	rpm
	  Dust Discharge  to Screw
                                                                           amps
     Liquid level indicator  Yes
     Liquid temperature 	
     Flow in	gal/min
     Percent solids in	
     Instrumentation:
     Primary Current 	
     Primary Voltage 	
     Secondary Current 	

     Secondary Voltage 	
                                               No
                                        Flow out
                                        Out
            gal/min
                                             Gas Inlet Temperature
                                             Level  Indicator Alarm
                                             Hopper Heater Indicator 	
                                             Light
                                             Penthouse Heater Indicator
                                             Light
     Shell
     Material of Construction
     Insulation:  Yes
                                                       No
     Opacity Monitor
     Number
          Manufacturer
          Model Number
          Location
          Power Supply
          Type	
          Pulse System:  Yes _
          Digital Control Mode
          Other
                                                  No
 Figure 2-4.  Electrostatic precipitator inspection data sheet.   (Continued)
                                     48

-------
LOCATION 	
DESIGNATION 	
INSPECTOR(S) 	
CLAIMED
CONFIDENTIAL Yes
No
                     DATA SHEET NO.
                     DATE
     DESCRIPTIVE INFORMATION
     Fabric Filter Type 	
     Manufacturer 	
     Model No.
     Cloth Type and Weight
     Area
     Air-to-cloth Ratio
     Pressure Drop Range 	
     Operating Temperature
     Collection Efficiency
     Type of Bag Cleaning:
     Shaker
            ft2
            ACFM/ft2
            in. H20
       Pulse Jet
Reverse Air
     Frequency of Bag Cleaning
     Bypass System:  Yes 	
           No
B.   FABRIC FILTER LAYOUT SKETCH
       Figure 2-5.   Fabric filter inspection data sheet.   (Continued)

                                     49

-------
C.   COMPONENT INFORMATION
     1.
     2.
     3.
Hoppers
Number _
Length _
Shape
                                       ft
                                       degrees
Level Indicator Height _
Insulation:  Yes	  No
Heating:  Yes	  No __
Shell
Material of Construction
Insulation:  Yes	  No
Opacity Monitor
Number
                                                 ft from top
          Manufacturer
          Model  Number
          Location
      Figure 2-5.  Fabric filter inspection data sheet.  (Continued)
                                     50

-------
                         REFERENCES  FOR  SECTION  2
 1.  Nelson, W.  A  New Theory  to  Explain  Physical  Explosions.   TAPPI,  March
    1973.

 2.  Taylor, Malcolm  L.,  and Howard S.  Gardner.   Causes  of Recovery Boiler
    Explosions.  TAPPI,  November 1974.

 3.  Petri, H.  The Effects of Hydrogen Sulfide  and Carbon Disulfide.   Staub
    2i(2)-:64,  1961.

 4.  Denrnead, C. F.   Air  Pollution by Hydrogen Sulfide from a  Shallow Pollu-
    ted  Tidal  Inlet.  Auckland,  New Zealand.  Clean Air Conference, Uni-
    versity of New South Wales,  1962.

 5.  Permissible Emission Concentrations  of Hydrogen Sulfide.   Subcommittee
    on Effects of  Hydrogen Sulfide of the Committee on  Effects of Dust and
    Gas  of the Verein Deutscher  Ingenieure Committee on Air Purification,
    VDI  2107,  1960.

 6.  Patty, F.  A.   (Ed.).  Industrial Hygiene and Toxicology,  Vol. I.   (New
    York:   Interscience, p. 896, 1958.)

 7.  U.S. Department  of Health, Education, and Welfare.   Preliminary Air
    Pollution  Survey of  Hydrogen Sulfide.  A Literature Review.  Contract
    No.  PH 22-68-25, October  1969.

 8.  Adams, D.  F.,  and F. A. Young.  Draft Odor  Detection and Objectionability
    Thresholds.  Washington State University Progress Report on U.S. Public
    Health Service Grant, 1965.

 9.  U.S. Department  of  Health, Education, and Welfare.   Preliminary Air
    Pollution  Survey of  Chlorine Gas.  A Literature Review.  Contract No.
    PH 22-68-25, October 1969.

10.  Heyroth,  F.  F.  Chlorine, CL2, in Industrial Hygiene and Toxicology,
    Vol. II,  2nd  ed.  (New York:  Intersicence, 1963.)

11.   Kowitz,  T. A., et al.  Effects of Chlorine Gas upon Respiratory Function'.
    Arch.  Environ. Health 14:545.  1967.

12.  Manufacturing  Chemists Association.   Chemical Safety Data Sheet SD-80,
     Chlorine.   Washington, D. C.  1960.

13.  The Chlorine Institute,  Inc.  Chlorine Manual.  New York.  1959.

                                       51

-------
14.  Beck, H.  Experimental Determination of Olfactory Threshold for Some
     Important Irritant Gases (Chlorine, Sulfur Dioxide, Ozone, Netrosyl-
     sulphuric Acid) and Their Manifestations of the Effect of Law Concen-
     trations in Maa, Thesis, Wurgburg.  1959.

15.  Takhirov, M. T.  Determination of Limits of Allowable Concentrations of
     Chlorine in Atmospheric Air.  U.S.S.R.  Literature on Air Pollution and
     Related Occupational Dieseases 3:119.   1960.

16.  U.S.  Department of Health and Human Services.   NIOSH/OSHA Pocket Guide
     to Chemical Hazards.  September 1978.
                                     52

-------
                                  SECTION 3
                           KRAFT PULPING PROCESSES
     There are five major processes and several minor processes within the
kraft pulp mill.  The major processes are wood handling, pulping, chemical
recovery, causticizing, and power generation.  The minor processes include
bleaching and raw material handling.  Figure 3-1 is a simplified diagram of
the kraft pulping process.   This section presents descriptions of these
processes and discussions regarding the sources of emissions, the control
techniques, the potential malfunctions, and the inspection techniques to be
used for each process.

3.1  WOOD HANDLING DEPARTMENT
     This subsection describes the processes used to receive, prepare, and
transport logs to the pulping department.  These processes include debarking,
chipping, screening, transfer, feed preparation, and storage.  Most of the
emissions in the wood handling department are fugitive.   The control  measures
generally consist of containment and water.
     The major malfunctions generally involve the pluggage of the chip trans-
fer system and discharge of the chips from the cyclone exit.  The inspection
of the wood handling department consists of a visual evaluation of the fugi-
tive dust sources and a verification that the prescribed containment has not
been removed.  Inspections of the wood handling department are generally
Level  I Inspections.
3.1.1  Process Description
     The major processes within the wood handling department are described
briefly.
3.1.1.1  Debarking—
     Depending on the location of the pulp mill with respect to the wood
supply, wood is transported to the mill by railcar, truck, or barge in the
                                      53

-------
                             WOOD CHIPS
   EVAPORATOR
     GASES
                    STEAM
      i
1
        Figure 3-1.   Kraft pulping  process.
                            54

-------
form of chips, sawdust, shavings, or roundwood (logs).  Roundwood, generally
in four to eight foot lengths, is usually debarked before it is used in pulp
manufacturing.  Prior to the debarking, various techniques can be used to soften
the bark.  These include:
     o    Spraying the logs with water
     o    Soaking the logs in ponds
     o    Steaming the logs in special chambers
     After the logs are softened, they must be slashed or cut into 4 to 8 foot
lengths.
     Because bark has very little useful fiber, and it contains dirt that re-
duces the overall pulp quality, debarking is used to improve the quality of
the pulp.  Several types of debarking machines are commonly used:  drum, bag,
ring, cutterhead, hydraulic, and knife debarkers.   One of the most widely
used is the drum debarker (Figure 3-2).  The logs are fed into an open-ended
horizontally rotating cylinder shell that is 8 to 16 ft in diameter and 22 to
75 ft long.  The tumbling action of the logs removes the bark by abrasion.
The logs move toward the discharge end during the process, and the debarked
logs are expelled.  The bark removed from the logs falls through slits located
along the entire length of the cylinder.  Inadequately debarked logs are re-
turned for reprocessing.
     The debarking drums can be constructed of either staved sections or en-
closed vat sections.  Staved sections promote tumbling of the logs against
each other, which removes loose pieces of bark.  The enclosed vat sections are
used to soak the logs in a pool of warm water to thaw the bark at mills located
in areas where freezing temperatures are common.  If the bark is to be used
as boiler fuel, presses are often used to remove the moisture.
     Figure 3-3 depicts a bag debarker that abrades the bark from the logs by
jostling and rotating.  The removed bark falls through holes in the base of
the machine, and operation can be batch or continuous, wet or dry.  Water is
frequently used, to thaw frozen bark.
     Ring debarkers, shown in Figure  3-4, remove bark by a scraping action.
Individual logs are fed axially  through a rotating ring having equally spaced
arms held under pressure.  The removed bark is either flung from  the machine
                                         2
or allowed to fall onto a conveyor  below.

                                      55

-------
                        Figure 3-2.   Drum debarking.
Source:
                Figure 3-3.  Bag debarker.

Joint Textbook Committee of the Paper Industry, the Pulping of Wood,
1969.  Reproduced with permission of McGraw-Hill.
                                      56

-------
 Source:
                 Figure 3-4.   Ring barker.
Joint Textbook Committee of the Paper Industry, the Pulping of Wood,
1969.  Reproduced with permission of McGraw-Hill.
      Figure 3-5 shows a  cutterhead machine.   Bark is  removed  from one  log  at
 a time by the milling action of the cylindrical  cutterhead.   Cutting edges
 are mounted on the longitudinal axis,  which rotates parallel  to the axis  of
 the log.   The removed bark is carried  away by a  conveyor.
      Hydraulic barkers,  depicted in Figure 3-6,  blast the bark from the log
 by directing a high-pressure water stream at the surface.   The logs are de-
 barked one at a time.  The water then  washes away the removed bark.
      Figure 3-7 shows a knife debarker that uses radially mounted knives to
' strip bark and wood from the surface of the log.  The knives are mounted on a
               2
 rotating disk.
      If the debarked logs have diameters too large to enter the feed spout of
 the chipping equipment, they are  sent to a splitter that quarters them.  De-
 barked logs that  are not needed immediately are  stored for later  use.

  3.1.1.2  Chipping—
      After the logs  have been  debarked,  they must  be  reduced  in  size  so that
  cooking  chemicals can easily penetrate  the wood fibers to  separate lignin  and,
  carbohydrates  from the cellulose.  Chippers, which will  accept 4- to  8-foot

                                       57

-------
Source:
              Figure 3-5.  Cutterhead barker.

Joint Textbook Committee of the Paper Industry, the Pulping of Wood,
1969.  Reproduced with permission of McGraw-Hill.                   '>-
                                   58

-------
                      Figure 3-6.   Hydraulic barker.
                         Figure 3-7.  Knife barker.

Source:  Joint Textbook Committee of the Paper Industry, the Pulping of Wood,
         1969.  Reproduced with permission of McGraw-Hill.

                                     59

-------
  debarked logs, use powerful high-speed rotating knives to reduce the wood to
  a uniform size (about 5/8 to 3/4 inch long by 1/8 inch thick).
       Among the variety of chippers used,  the most common is the Norman disk
  chipper (Figure 3-8).   The close spacing  of the knives holds the log steady
  and prevents bouncing  in the feed spout.   In operation,  each knife moves  through
  a section of the log in the same direction as its bevel.   The logs are cut
, against the grain at an angle of 35 to 52 degrees relative to their axis.   At
  every point along the  cutting edge,  the bevel  is  directed  toward the edge  of  •
  the next chip  slot.  Every point along the knife  edge  exerts  the same force
  in  pulling  the log into the disk.   This technique reduces  the bruising  of  the
  chip fibers  and  results in a  stronger  pulp.
       Drum chippers  (Figure 3-9)  consist of an  open-ended drum that  rotates  at
  20  to 40 rpm.  The drum chipper  is fitted  with several equally spaced cutting
  knives on the  periphery that cut parallel  to the  grain of  the log.  The cut-
  ting assemblies consist of front knives with three cutting edges that cut      '.
  trapezoidal grooves in  the wood to form chips.  The remaining wood is sliced   ;
 off by a trailing knife to form additional chips.
      The spiral horizontal parallel (HP) chipper  (Figure 3-10) cuts parallel   ;
 to the grain.  Wood is fed into the disks mounted on a horizontal shaft
 arranged in a V shape.   Two or more rows of knives are mounted on each disk
 in a spiral  formation.
 3.1.1.3  Knotting and Screening--
      After passing through the chipper, the wood contains fines,  slivers,  and
 oversized chips.   A screening operation is performed to separate rejects and
 fines from properly sized chips.   Fines are generally less  than  1/8 inch in
 diameter and are usually pulped separately for low-strength paper,  or are
 burned in the bark boilers.   Oversized  chips  are separated  and removed for
 rechipping or recrushing and rescreening.
      Rechipping often damages  the chips and results in  increased  fines.
Several  types of rechippers are currently  being used, but the  chips  receive
less  damage  if  the rechippers  use  knives rather than  blunt  hammers.
      Screening  generally involves two decks.  The  upper deck captures  the
oversized chips and the  lower deck  retains  proper  sized chips.  Fines  drop
into a collection  hopper below  the  lower deck.  The kinds of screening equip-
ment, classified according  to movement  mechanism,  are rotary-drum, shaker,

                                     60

-------
                                           SE6MENTAL KNIFE HOLDER


                                                             -DISK
            LOG
                                                        CLAMPING BOLT

                                                       CHIP SLOT
                                                         STEEL INSERT

                                                        CHIP SLOT
            SPOUT


            PULL-
            ANGLE
            BEDKNIPE
             FRONT FRAME
                     Figure  3-8.   Norman disk chipper.
         s
Source:   Reproduced with  permission of  Carthage Machine Company,  Inc.
                                       61

-------
                         Figure 3-9.  Drum chipper.

Source:  Reproduced with permission of Koehring-Waterous Ltd.
Source:
        Figure 3-10.   Horizontal parallel chipper.

Reproduced with permission of Joint Textbook Committee of the Paper
Industry, the Pulping of Wood, 1969.                               '

                            62

-------
vibratory, and gyratory.  Vibratory and gyratory are the most efficient and
the ones most often used.  Vibratory screens (Figure 3-11) are inclined at
approximately 25 degrees from the horizontal;f  Gyratory screens (Figure 3-12)
are inclined at 4 degrees from horizontal, and they use r-ubber balls that
deflect against the underside of the wire mesh deck to dislodge chip particles
                       27
that plug the openings. *
3.1.1.4  Storage and Transfer-
     Chip feeders operating at a constant and uniform rate are used to move
the chips from storage.  Various equipment is used to accomplish this task,
including a traveling screw, a rotary plate, a stoker, a multiple screw, and
parallel drag chains.   Each has its own technique for moving the chips.
     Pneumatic conveying is frequently used when the chips are stored outside.
A pneumatic,system consists of a pipeline with a high-velocity airstream that
is generated by a blower.  The chips are fed into the airstream and transported
to the  desired point of exit.
     Chips, shavings, and sawdust arrive at the mill in freight cars, semi-
trailers, or trucks and must be removed to  storage areas  for future use.   Some
of the  techniques used  to transport the chips, shavings,  and sawdust are:
     o    Chip digger with suction unloading  (for boxcars)
     o    Rotating nozzle suction unloading  (open-top  cars)
     o    Hydraulically operated truck dumper
     o    Rotary  freight car dumper with  gravity-dump  pit
     o    End-dump freight car with pit.
     Once the  chips  are removed  from  their transporting vehicle,  they  can be
 transported to bins,  outside piles, or silos  by belt conveyor,  bucket  elevators,
 or pneumatic systems.   Usually the  bins  are loaded by overhead conveyors and
 unloaded by a  slot at the bottom that uses a  traveling screw.
      Pnuematic systems use cyclones  as primary receivers to separate chips
 from the transport gas stream.  The cyclones are 8 to 14 ft in diameter and
 typically operate at gas volumes of 4000 to 11,000 acfm.  The inlet velocities
 are on the order of 100 to 160 ft/s.   In this application, the cyclones are
 used as product transfer and are not considered an air pollution control de-
 vice.   Typical transport rates are 50 to 100 tons/h.-  The cyclone may be
                                      63

-------
                    "OS" SPRING SUSPENSION
                    ASSEMBLIES
                         FEED PLATE
                                                         IIUBBEM COVERED SUPPORT MAILS


                                                                     TENSIONINO STRIPS
            "K" SPRING FLOOR
            MOUNTING WITH STABILIZERS
                                                                 •CUttN END SUPPORT
                                             DtSCHAMQE PANEL LIP
                         Figure 3-11.   Vibratory chip screen.

Source:   Reproduced  with  permission of W.  S. Tyler  Company.
                                             64

-------
       INTAKE
                                           onrr, SAWDUST
                                          AND OTHER FINES
                                                        USABLE OOPS
                                                        DISCHARGE OF
                                                        OVERSIZE CHIPS
                     Figure 3-12.  Gyratory  chip  screen.
Source:  Reproduced with permission of  ROTEX,  Inc.

used to transfer the material  to  screens,  the  hammer mill,  or the chip storage
bins.9
     Figure 3-13 shows a typical  end-dump  rail  car unloading system, and Figure
3-14 shows a hydraulic truck  dump system.  Chips  received in rail cars can be
unloaded by a rotary rail car dump (Figure 3-15).
     Figure 3-16 shows the normal  material and air flow at a small wood-
              0
handling yard.
3.1.2  Sources of Emissions and  Control
     Most of the emissions from  a woodyard,  except for those from pneumatic
conveying systems, are fugitive.   In  general,  control measures consist of con-
tainment of sources and  the use  of water on  haul  roads and other traveled
areas.  Water may also be used on the debarkers to reduce dust and to wash the
logs.
     Chips that are received  dry (shaving and saw dust) are also potential
fugitive sources.  Water is effective in reducing the emissions at transfer
points.
                                      65

-------
                Figure 3-13.  Hydraulic dump truck for chips.
Source:  Reproduced with permission of Screw Conveyor Corporation,
                                      66

-------
    Figure 3-14.   End-dump freight car and unloading platform for chips.

Source:  Joint Textbook Committee of the Paper Industry, the Pulping of Wood,
         1969.  Reproduced with permission of McGraw-Hill.
                                      67

-------
              Figure 3-15.  Rotary freight car dryer for chips.

Source:  Joint Textbook Committee of the Paper Industry, the Pulping of Wood,
         1969.  Reproduced with permission of McGraw-Hill.
                                     68

-------
                                                   EXIST.
CTl
ID
                               EXIST.
         EXIST. TRUCK « RAIL CAR
          CHIP UNLOADING SYSTEM
         E~JE~3
        -Jft   tMIIMi     M
                600
             TONS: SHIR
 RING SHOWER       I  OVERSIZE
(WOOD WASHER)  CONV.
                       EXIST.
                       BLOWER
                                       TO MILL
                                      50 TONS/HR
                                   (EMERGENCY ONLY)
                                                                                                     FINES TRUCKED
                                                                                                      TO LANDFILL
                         Figure 3-16.   Typical material and  air flow for small woodyard.

-------
      In urban areas, more control  is needed because of the potential nuisance
to surrounding property owners.  As a result, several plants within urban areas
have  paved the haul roads and chemically stabilized or sprayed the storage
areas.  Periodic sweeping and/or wetting of these haul roads has been effec-
tive  in reducing emissions resulting from vehicular traffic.  Minimiaing the
dust  around the plant side has the added benefit of reducing the carryover of
dust  and dirt into the pulping process.
      Emissions from the pneumatic conveying cyclone are generally controlled
by the use of water sprays.  These water sprays can reduce cyclone emissions
by 95 percent.  Grain loadings from a chip transfer cyclone (14 foot diameter
and 10,000 acfm) handling 100 tons/h of dry chips can range from 0.0003 to
0.034 gr/dscf.  The lower value is associated with a cyclone using water sprays.
A typical water injection rate is approximately 4.5 gpm (0.5 gal/1000 acfm).
Green chip transfer through the same system generate an emission rate of 0.002
gr/dscf.9
3.1.3  Malfunctions
     Malfunctions that can occur in the wood handling department involve the-
pluggage of the chip transfer systems and the discharge of the chips from the
cyclone exit.  These malfunctions usually cause a fallout problem on the plant
premises, which normally can be corrected quickly.
     When water sprays are used to control  fugitive dust from the transfer and
debarker systems, provisions must be made to prevent freezing during the
winter months.
3.1.4  Inspection of Wood Handling Department
     The inspection of the wood handling department consists of a visual evalua-
tion of the fugitive dust sources and a verification that the prescribed con-
tainment has not been removed;  If water spray systems are used, the inspector
should verify the location of spray nozzles and visually determine if the
water spray pattern is adequate.
     The water flow rate should be recorded along with the water supply pres-
sure for each system.   Table 3-1 provides a checklist for process weight and
control variables of the emission sources in the wood handling department.   '.
Generally the information listed in Table 3-1 can be obtained during a Level:
I or  II inspection.
                                     70

-------
   TABLE 3-1.  CHECKLIST FOR INSPECTION OF WOOD HANDLING SYSTEMS
Debarker:




Chipper:




Screens:



Rechipper:


Truck unloading:
Type
Process rate
Control method_
Water flow rate
Water pressure _
Type	  Process rate
Control method 	
Water flow rate 	
Water pressure 	
Type.	Process rate
Reject rate 	
Control method 	
        •   Process rate
Control method 	
Type	Process rate
Control method  	
Rail car unloading: Type
           Process rate
                    Control method
Pneumatic transfer: (more than 1 repeat)
Process from
Process rate
Gas volume
Cyclone diameter
Inlet deminsions
to
tons/h
acfm
ft
X
 Inlet area	  ft
 Water spray flow rate 	
 Duct leaks 	Yes  	No
                                          in.
                    gpm
              _gpm
               psig
            _gpm
             psig
                                 71

-------
 3.2  PULPING DEPARTMENT
      The pulping  process  involves  many  separate  processes,  including the
 cooking  of the chips  to produce  pulp, recovery of cooking chemicals, and con-
 centration of spent liquors.   For  simplicity, all of  these  processes are in-
 cluded in this subsection.
      The primary  emissions  from  the  pulping department are  TRS compounds which
 can be controlled by  1) reduction  in quantity through process operation,    !
 2)  scrubbing with fresh water  or alkaline solutions, or 3)  incineration.  The
 malfunctions in the pulping department  involve the digester relief systems,
 digester blow system  (hot water  accumulator and  blow tank), multiple-effect
 evaporator system,  black liquor  oxidation system, and the TRS Vent Control  ;
 System.   In general,  all the major malfunctions  associated  with the above
 systems  will  tend to  increase  TRS emissions in one way or another.  Several
 tables are presented  in Section  3.2.3 that identify the potential malfunctions
 for each type of  system, the primary effects of  these malfunctions, and the
 subsequent results  in terms of TRS emissions.
      The inspection of  the  pulping department is generally  based on information
 concerning equipment  specifications, process procedures, process weights, and/
 or  control  equipment  bypass and malfunction.  Because the inability to obtain
 physical measurements with  hand  held equipment in the pulping department, in-
 spections  in  the  pulping department  are generally limited to Level I or II.
 A series of check lists are presented in Section 3.2.4 that will aid the in-
 spection in  obtaining whatever information is available for the major systems
 in  the pulping department.
 3.2.1  Process Description
 3.2.1.1  Pulp Digester—                                                    ;
                                                                        «
     The two major components of pulpwood are cellulose and lignin.  The fibers
of cellulose are bound  by the lignin in the wood.  Figure 3-17 presents a
schematic of the pulping process.  Wood chips are added to a reaction vessel
 (digester) containing cooking chemicals comprised of about 66 percent sodium
hydroxide  (NaOH) and 33 percent sodium sulfide (Na2S).  Some recycled black
liquor is also added to the vessel.  The mixture is heated to 340° to 350°F
under 100 to 135 psig of steam for lh to 3 hours.  This process removes the
lignin and carbohydrates from the pulp to free the cellulose.   Sodium

                                      72                                    ;

-------
CO
                                                                     SECONDARY
                                                                     CONDENSER
                                                                                  11  HOT WATER
                                                                                  10  COOLING WATER


                                                                                  (5) RELIEF GAS
                                                                                 (8) TURPENTINE


                                                                                 © FOUL CONDENSATE

                                                                                 (?) BLOW GAS
                                                                        HEAT
                                                                        EXCHANGER
                                                                                  5  HOT WATER
                                                                                  A  WARM WATER
                                                                                 ©  BLOW CONDENSTATE BLEED

                                                                                  2  FRESH WATER
                                                                                 (?)  PULP-LIQUOR
                               POINTS OF  POSSIBLE ODOR RELEASE ARE ENCIRCLED,
                                     Figure  3-17.  Pulping process.
                                                                  12

-------
hydroxide breaks down the fibers in the wood.  Sodium sulfide, by double de-
composition reaction, breaks down to liberate NaOH as it is consumed.  Thus
Na2S acts as a reservoir for NaOH so that excessive amounts are not added,
which would weaken the pulp.  Sodium sulfide is responsible for the strength
attributable to kraft pulp.
     The digestion process can be either batch or continuous, and the digester
can be heated directly or indirectly.  The digester generally has a volume of
between 3000 and 6500 ft3 and will normally produce between 8.5 and 17.9 tons
of unbleached air-dried pulp per blow.  The digesters are charged with white
liquor, chips, and recycled black liquor.  Table 3-2 shows the typical liquor
                                              O
requirements for bleached paper and bag paper.   The liquor requirements, pulp
yield, pulp strength, and cooking time are mill-specific.

             TABLE 3-2.  TYPICAL DIGESTER LIQUOR REQUIREMENTS2
Pulp type
Bleached papers
Bag paper
Total liquor per
blow, ftVTADP
100 - 235
80 - 170
White liquor per
blow, ftVTADP
80 - 120
65 - 115
Black liquor per
blow, ftVTADP;
5 - 110
5-90
     A typical digester temperature/pressure chart for repeated digester batch
cycles is shown in Figure 3-18.2  The total number of potential digester blows
per day may be calculated from the number of digesters and the average digester
cycle time, including charging, cooking, and blowing.
     Chips are charged to the digester from overhead chip bins in the pulp  i
building.  Minor steam is lost and odors are generated during charging.  Re-
movable charge chutes are used to transfer the chips.  Figure 3-19 shows a
typical digester charging room floor.
3.2.1.2  Batch Digester (Blow)—
     During batch digestion, the pulp is held at elevated temperature and
pressure.  To remove the pulp from the tanks, the digester bottom is opened
and the tank pressure is relieved to a blow tank.  The process is referred to
as a digester blow.                                                         ;
                                     74

-------
              Figure 3-18.  Typical digester operating curves.

Source:  Joint Textbook Committee of the Paper Industry, the Pulping of Wood,
         1969.  Reproduced with permission of McGraw-Hill.
                                     75

-------
             Figure 3-19.  Typical digester charging room floor.
Source:  Joint Textbook Committee of the Paper  Industry, the Pulping of Wood,
         1969.  Reproduced with permission of McGraw-Hill.
     Because the temperature and pressure of the digester contents are quite
high, the rapid release of pressure flashes a portion of the water to steam.
The amount flashed depends on the blow rate and the initial and final temperature.
A large pressure vessel (blow tank) is provided for separation of the pulp
from the spent cooking chemicals (black liquor) and to allow expansion volume
for the evolved steam.  For economical operation of the digester system, the
blow tank is equipped with a hot water accumulator (Figure 3-20), which may
have both primary and secondary condensers to recover heat from-the flashed
steam.  The primary condenser is generally a direct-contact type and the
secondary condenser is a tube and shell design.
     Calculation of the exhaust volume from the blow can be based on liquor
volume, system mass balance, and specific heats of the materials.  After con-
densing, the gas volume is much lower, and the gas contains equilibrium moisture
at 212°F and 1 atm plus any noncondensable gases and trapped air.

                                     76

-------
                              BLOW    SECONDARY
       BLOW
       STEAM

     PRIMARY
   CONDENSER
   BLOW
CONDENSATE
                               GAS
                                    CONDENSER
                                                                 HOT WATER
                                                              4  WARM WATER
                                                                COLD WATER
             Figure 3-20.  Hot water accumulator showing primary
                         and secondary condensers.12
     The digester blow is a significant air pollution source of TRS compounds.
The reaction of the wood with white liquor sulfide generates hydrogen sulfide
(H2S), turpentine, and traces of methanol (CHgOH), ethanol (CHgCHgOH), and
acetone (CH2COCH3).  The composition and quantity of digester gases will
differ between batch digesters and continuous digesters.  Generation rates will
also differ depending on wood type, white liquor sulfidity, final cooking
temperature, and cooking time.
     The typical blow volume is 90 percent steam.  The noncondensable portion
                                 3             12
of the blow volume is about 95 ft /ton of pulp.    The blow rate is not con-
stant and reaches a peak during a portion of the blow period.  Figure 3-21
shows a typical digester steam flow pattern.    The primary and secondary
condensers are designed for the peak flow rate.
     The digester blow volume may be calculated from the digester operating
conditions and a liquor mass balance.  Most of the information required to
complete the calculation is available from the pulp cooking records and/or
digester design data.  At the end of the cooking period, the digester is
                                     77

-------


      100 -
      90 -
      80 -
      70-
      50 '
   LU
      40 -
   w
   5  30 -
   §  20-
       10 -
       0
 h 2500 I
                                                                       »_
 L 2000

 - 1500

   1000

   500
                           10               20
                           TIME, MINUTES
30
     Figure 3-21.  Typical batch digester steam flow rate during  blow.
                                                                      13
discharged into the blow tank and the temperature drops from 350  to  220°F.
Some steam is flashed during the blow period.   The basic work sheet  for  cal-
                                                 2
culating blow weight is presented in Figure 3-22.   The calculation  represents
a balance between total heat available from the liquor and total  heat availa-
ble from the pulp and digester shell.
3.2.1.3  Digester Relief and Turpentine Recovery—
     The digester cooking cycle generates organic gases as a result  of volati-
lization of wood oils.   For maintenance of acceptable  cooking pressures, the
digester must be periodically vented to the atmosphere.   The relief  gases are
passed through a liquor separator (cyclone) to remove  any entrained  liquor
and fiber (Figure 3-23).     From the separator the gases are passed  through a
surface condenser» which removes the condensable gases (primarily turpentine
and water).   The condensate is passed to a decanter, where the water and
organics are allowed to separate.   Crude turpentine is drained off to storage-
                        2 12 14
and shipped to refiners.  *   '                                              ;
                                    78

-------
Weight and specific heat of black liquor discharged:

     Lignin, etc.
     Water in chips
     Cooking liquor
     Black liquor
                            15,800  Ib
                            26,200  Ib
                            59,000  Ib
                            30.000  Ib
                                                  Total     131,000 Ib

     Relief vapor                                           -1.850 Ib

     Total weight of black liquor discharged               129,150 Ib

     Liquor specific heat                                    0.855

Weight and specific heat of moisture-free pulp discharged
     Pulp weight
     Pulp specific heat

Weight and specific heat of digester shell

     Shell weight
     Shell specific heat

Mass balance - Blow heat available, 10  Btu
     Black liquor
     Pulp
     Heat from digester shell
(129,150)(0.855)(350-220)
(  14,400)(0.33)(350-220)
(  63,270)(0.117)(350-255)
          Total heat

Latent heat of steam at 220°F

Weight of steam flashed (excluding 1t- /-on nnn
  noncondensable gases) at 220°F = 1a'g°"'uuu
      965
                            14,400  Ib
                              0.33
                            63,270  Ib
                             0.117
= 14.355
=  0.620
=  0.705

= 15.680

     965


= 16,250 Ib/blow
 Figure 3-22.  Worksheet.for calculation of blow weight (steam).
                                79

-------
           RELIEF COMPUTING
           AND CONTROL SYSTEM
                                                   DECANTER
                STRAINER
                  LIQUOR
                  SEPARATOR
                                                   CONDENSATE

                                         CRUDE TURPENTINE
             BLOW VALVE
            	JXJ	

             CONTROL SYSTEM OF TURPENTINE COLLECTION
Figure 3-23.  Digester relief and turpentine recovery system.15


                            80

-------
     Typical turpentine yields from soft wood pulping are on the order of 2
to 4 gal/TADP.   The turpentine yields from hardwood pulping are minimal  and
recovery is not practiced.   Continuous digesters also generate turpentine as
a result of continuous relief during pulp blowing.   The yield and quality of
turpentine from continuous  digesters is generally inferior.
     The amount of condensate produced by softwood pulping is mill-specific,
but is generally about 540  Ib/ton of pulp.  The volume of noncondensable gases
                                 3             12
produced is on the order of 32 ft /ton of pulp.
     The amount of noncondensable organic emissions present in the condenser
vent is a function of the condensate outlet temperature.  Figure 3-24 shows
the relationship between TRS emissions and condenser outlet gas temperature.
                                                                            17
3.2.1.4  Continuous Digester--
     Continuous digestion presents less of an air pollution problem than batch
digestion because the gas flow is regular.  Spent liquor from the digester is
expanded (flashed) in two stages.  The flash from the primary tank is used to
impregnate chips in a presteaming vessel prior to digestion.  Steam from the
secondary flash tank is used for various purposes, such as turpentine recovery,
hot water production, additional impregnating steam, etc.  In both batch and
continuous digestion when soft woods are digested, the condensate undergoes a
secondary process that separates the turpentine from water.  Figure 3-25 depicts
             12
this process.
     The water that condenses from the digestion process is used for brown
stock (pulp) washing.  Prior to being washed, however, the pulp goes through
a knotting operation that further dilutes the fiber to lh percent by weight
and removes oversized chips that have not been completely impregnated with
white liquor.  These chips are sent back through the digester, and the water
is drained into filtration tanks.
3.2.1.5  Pulp Washing--
     After cooking, the pulp contains a considerable quantity of chemicals
and dissolved waxes, oils, and fatty acid salts.  These  chemicals are re-
coverable when converted to their original form through  the recovery process.
     Pulp from the blow tank is mixed with white liquor  and black liquor  in
the blow tank and transferred to the pulp washing lines.  The pulp washing
lines are of the displacement-washing type.   In this process the liquor and
                                     81

-------

20 •
"E
^ 15-
to
CD
z
o 10 -
|
H
0
8 5-
n -
* .
o
o

• ° x
X *
•
X
X
X* ,
0
o o
i
A 0 0 °
         50      60       70       80       90

           CONDENSER  OUTLET TEMPERATURE,  °C

             •     HYDROGEN SULFIDE
             o     METHYL  MERCAPTAN
             X     DIMETHYL SULFIDE
             A     DIMETHYL DISULFIDE
 Figure  3-24.  Odor compounds in relief gas after  turpentine
condenser as a function of condenser outlet temperature.10 '
12,17
                          82

-------
14 STEAM
15  COND.
                     PRESTEAMING  if"5^
                     VESSEL
13 WHITE LIQUOR



12 CHIPS


©  VENT

11 HOT WATER
10 COOLING WATER



(§) TURPENTINE




(7)  FOUL CONDENSATE


l6  FLASH STEAM
                                                            (|0)  WEAK LIQUOR

                                                         ^   ©  PULP + LIQUOR


                                                            OT)  WASH LIQUOR
             POINTS OF POSSIBLE ODOR RELEASE ARE ENCIRCLED
               Figure 3-25.   Continuous digester flow  sheet.12
                                       83

-------
 chemicals  are displaced in  the pulp by fresh  water.   The washing  is accomplished
 in  stages, with waste liquor being  transferred  countercurrent to  the pulp flow.
 The washing usually takes place on  rotary drum  filters, and air is used to
 maintain a pressure differential  over  the washed  pulp sheets.
      The hot black liquor is exposed to large quantities of air and generates.
 large volumes of steam.  The exposure  of the  liquor to ambient air has two
 effects:   1)  the reduced sulfur in  the black  liquor is partially  oxidized,
 and 2)  a portion of the TRS  compounds  is flashed  through the washer vents.
      Three types of pulp washers  are commonly used:   vacuum washers, pressure
 washers, and  diffusion  washers.   In  the vacuum  washer, the most common (Figure
       12
 3-26),   the  washer system has  three sections:  a drum, liquor tanks, and a
 foam  tank.  The foam tank and drum,  which  are vented  to the atmosphere, are
 sources of TRS  emissions.  The  drums are typically hooded in groups and vented
 directly through the mill building  roof.   The volume  of gases exhausted from
 the drum hood is substantial  (48,000 to 190,000 acfm/TADP) and the concentra-*
 tion  of TRS gases  is  low (25  to 30  ppm).12
      The pressure  washer is  similar  to  the vacuum washer, except  that the     i
 drums are  enclosed with recirculating  pressure  blowers (Figure 3-27).  The
 quantity of air vented  from  the drum is  smaller (less ambient air exhausted
 from  the building),  but the TRS levels  in  the gases are higher.   Because the
 gas volume  is smaller, TRS emissions from  pressure washers can generally be
 controlled  by incineration.
      Diffusion  washing usually  is accomplished  in a closed reactor (Figure
 3-28) without air,  which reduces  liquor oxidation and odor emissions.  The
washers may be  batch  or continuous.
      Black  liquor  recovered from  the washers  is generally between 10 and 15
 percent solids  by weight and  is only partially  oxidized.   The liquor also
 contains recoverable  amounts of tall oil soap, waxes, and fatty acids.
     The washer  lines contain multiple  stages that have specific design pulp
process weights  and  liquor flow rates.    Operation at excessive pulp rates can
result in incomplete  chemical removal from the pulp and/or reduced chemical
recovery through the washing liquor.
3.2.1.6  Black Liquor Concentration  (Evaporation)—
     The most common method of concentrating the black liquor from the brown  '
stock washers is indirect contact with steam.   Most sources prefer a multiple-
effect evaporator  (MEE), which is generally composed of three to seven  stages.
                                     84

-------
              HOOD
                                                        VENT
—*[
                                                     4  WASH
                                                     3  WASH
                                                     2 WASHED PULP
                                                       PULP
                                                        VENT
                                                    QO) WEAK
                                                        LIQUOR
POINTS OF POSSIBLE ODOR RELEASE ARE ENCIRCLED
         Figure 3-26.  Vaccum washer flow sheet.
                                                  12
                             85

-------


1
r
'

LIQ.
TANK





1

,r
LIQ.
TANK






1


«

LIQ.
TANK

t
r
^
c
t
r
ft*
FOAM
TANK

-«-»•
                                                   (5) VENT


                                                   4  WASH
                                                   3  WASH

                                                   2  WASHED PULP


                                                   (?) VENT
                                                   GO)  WEAK LIQUOR
POINTS OF POSSIBLE ODOR RELEASE ARE ENCIRCLED
         Figure  3-27.   Pressure washers  flow sheet.
                                                     12
                                86

-------
1
1
! i
DIFF.
WASH
1





! i
DIFF.
WASH
2
1


•«•••
I
i
i 4r
DIFF.
WASH
3
^. 9 UIACU
..fe. (Toi WFAK
       BATCH DIFFUSION WASHERS FLOW SHEET
                                   (5) VENT





                                     WEAK LIQUOR





                                   3 WASH





                                   2 WASHED PULP
                                    1 PULP + UOUOR
POINTS OF POSSIBLE ODOR RELEASE ARE ENCIRCLED.
  CONTINUOUS DIFFUSION WASHERS FLOW SHEET
  Figure 3-28.   Diffusion washer flow sheet.12
                      87

-------
                                           18
Each line in the evaporator is normally equipped with condensate flashing,
liquor preheating, hot water generating, degassing, tail steam condensing,
and vacuum generating equipment (steam ejectors).  Methods of liquor movement
can be falling film, rising film, or forced circulation.  The majority of the
effects are a long-tube vertical and a shell design.  As Figure 3-29 shows,
the most concentrated liquor is heated by high-pressure steam.   The resulting
steam is then used to heat the liquor in the next effect.  Each effect operates
at successively lower pressure, thus lowering the boiling point of the liquor.
     Evaporated water vapor from the last effect is condensed in either
barometric condensers (direct contact) or tube and shell noncontact (Figure
                                     12
3-30) condensers with steam ejectors.    The vapor removal rate of the con-
densers must be high enough to maintain a vacuum above the vapor space in
the last effect.  Noncondensable gases are vented from the tail gas condenser '
with a small steam ejector.
     Storage tanks are used between the middle effects (liquor solids 28 to 30%)
to allow decanting and skimming of tall oil soap from the concentrated liquors.
The size (volume/height) of the soap skim tanks is defined by the vertical
velocity of soap micelles in the liquor.  The vertical velocity is affected
                                            18 19
by liquor pH, temperature, and micelle size.  *
     The storage tanks and condensate hot wells, which are generally vented
to the atmosphere, are a source of TRS emissions.  Noncondensable gases
generated in each effect must be vented to maintain heat transfer coefficients
and to prevent corrosion of the shell.   The venting may be accomplished by
either single- or two-stage systems.  Single-stage venting consists of the
direct venting of each effect to a central manifold.  The gases are typically
vented to the secondary condenser, and noncondenables are released through the
                              12
hot well or steam jet ejector.
     In two-stage venting the first two stages after the liquor feed stage are
vented.  The noncondensable gases (5 to 15% of the total vapor flow) are
vented to separate condensers that are then vented to the main vacuum system.
     Black liquor generally enters the MEE line at 12 to 15 percent solids
and leaves the last effect at 48 to 55 percent solids.  Flow rate (gallons/
minute) and liquor weight through the effects are measured in total liquor
flow in and out of the line plus liquor percent solids in and out of the line.
88

-------
               VAPOR
    PROCESS
     STEAM
      CONDENSATE
                  THICK BLACK LIQUOR
                                           WEAK BLACK LIQUOR
Figure 3-29.  Multiple-effect, long-tube vertical evaporators
                       (backward feed).*
                               89

-------
   STEAM
 20
22
                                         '4^) Ub

                                     SEC'Y CONDENSATES
                                                                                            18  WARM WATER
                                                                                            17  FRESH WATER
                                                                                            16  FRESHWATER
                                                                                            14  FRESHWATER
                                                                                            15  STEAM

                                                                                               VENT
             THICK
             LIQUOR
 WEAK
LIQUOR
SOAP
   6+7+81)

TAILCONDENSATES
                         POINTS OF POSSIBLE ODOR RELEASE ARE ENCIRCLED.

                Figure 3-30.  Multi-effect, vacuum evaporation  plant flow sheet.12

-------
Normally, the steam flow to the effect is also measured.   Figure 3-31 shows a
typical temperature chart for liquor steam system.
     The amount of noncondensable gases vented from the hot well and tail  gas
condenser may vary from mill to mill, depending on the tightness of the system
and the amount of air inleakage into the system.20'21  Published data indicate
qas volumes between 10 and 400 ft3/TADP for hot wells.  A reasonable value
                            •                    3      12
for a southern softwood is on the order of 32 ft /TADP.
3.2.1.7  Condensate Stripping—
     Condensates from the digesters and evaporators contain odorous compounds
that may be lost to the atmosphere during water treatment and storage.  As a
result, many new mills operate condensate treatment plants to lower water
treatment biological oxygen demand (BOD) and reduce the potential loss to the
atmosphere.
     The organic compounds are divided into two classes, those  that are wood
oil bases containing no sulfur and those that are of  the TRS type.  The latter
class  generally has a very low odor threshold and is  maladorous.  Table 3-3
       •                                               22 23 24
lists  the main components of general mill condensate.  '   *
     The composition and amounts of condensates are a function  of process
operation,  pulp yield, cooking chemicals, cooking time, and dilution.  Table
3-4 presents species concentration data  obtained  from 10 mills.
     There  are several ways to control condensate odor problems:  reduction
in quantity, segregation of clean condensate, weak liquor  oxidation,  chlori-
nation,  and air or steam stripping.   Figure  3-32  shows a typical air
stripping condensate plant  flow  sheet.12 The  condensates  are acidified and
passed through a  separation column.   Overall TRS  removal efficiency can be as
                    12
high as 95  percent.
     Steam  stripping  is  similar, but the column is steam-heated and equipped
with primary and  secondary noncontact condensers  (Figure  3-33). The column
 is generally a  bubble-cap  type containing 10 to 20 trays.   '
     Removal efficiency is a  function of the steam-to-condensate ratio
 (Figure 3-34).   The pH of the condensate feed also  influences  H2S  removal
 rates.26'27  Noncondensable gases are vented from the condensers and usually
                                              28
 incinerated in  the lime kiln  or power boiler;
                                      91

-------
              Figure 3-31.   Chart of evaporator temperatures.

Source:  Joint Textbook Committee of the Paper Industry, the Pulping of Wood,
         1969.  Reproduced with permission of McGraw-Hill.
                                     92

-------
                                                             oo 03 04.
TABLE 3-3.  MAIN COMPONENTS OF TYPICAL KRAFT MILL CONDENSATES"'"'
No.
1
2
3
4
5
6
7
8
Component
CH3pH
CH3CH2OH
CH2COCH3
Turpentine
(Pinene)
H2S
CH3SH
CH3SCH3
CH3SSCH3
Boiling point,
°r
64.7
78.5
56.5
154
-59.6
7.6
37.5
117
(°F)
(148.5)
(173.3)
(133.7)
(309)
(-75)
(45.7)
(99.5)
(243)
BOD,
kg/ kg
1.00
1.23
0.67
-
0.60
0.07
0.31
0.61
Sulfur,
%
0
0
0
0
94
67
52
68
Odor threshold,
PPb
100,000
10,000
100,000
-
0.4-5
0.4-3
1-10
2-20
  TABLE 3-4.   TYPICAL KRAFT MILL CONDENSATE COMPOSITIONS, MEAN VALUES
                            FOR 10 MILLS23
Condensate
compound
H2S
CH3SH
CH3SCH3
CH3SSCH3
Total S
CH3OH
CH3CH2OH
CH3COCH3
Total BOD
Turpentine
decanter,
mg/liter
90
250
400
130
. 550
6,500
1,600
160
860
Means: Digester
blow, mg/liter
60
80
70
50
180
4,300
500
40
490
Evaporator
effects ,
mq/ liter
40
10 '
5
5
51
10,000
60
6
1,070
Evaporator
hotwel 1 ,
mq/ liter
100
40
7
15
135
1,000
40
10
1,060
                                   93

-------
    STRIPPING
    COLUMN
                                            8)VENT

                                            5) FOUL CONDENSATES
      STORE
      TANK
                            Jl
                    2 REST ACID
           L_r
                                            4 AIR
BLOWER
                                       •>   1 CLEAN CONDENSATES
          POINTS OF POSSIBLE ODOR RELEASE ARE ENCIRCLEa
                                                                      i ?
Figure 3-32.   Contaminated condensates air stripping plant flow sheet.
                                   94

-------
 SECONDARY  i
 CONDENSER  '
PRIMARY
CONDENSER
                    8 VENT


                    7 FRESH WATER

                    6 WARM WATER
                                          5 TURPENTINE PHASE
    HEAT
    EXCHANGER
                                          4  STEAM
                                          2  REST ACID
                                          1  CLEAN CONDENSATES
        Figure 3-33.
Contaminated condensates steam stripping
    plant flow sheet.12
                                 95

-------
o


§
o
U_
U_
LU
LJ
or
     100  -T
     80  -
     60
40  -
     20  -
                  2        4        6         8



                    STEAM/CONDENSATE, AS  %
                                                  10
 Figure 3-34.   Stripping efficiency for different steam-condensate

               ratios at 10 theoretical plants.12
                               96

-------
3.2.1.8  Black Liquor Oxidation—
     If a contact evaporation is used in the chemical  recovery process,  the
black liquor that is separated from the pulp during the washing cycle is nor-
mally oxidized with air to control the sulfide level and prevent the release
of odorous compounds.  As Figure 3-35 shows, the process consists of counter-
currently passing the black liquor through an air stream by means of a porous
diffuser,29 sieve tray tower,30 packed tower   or agitated air sparge.
The
first two techniques are more efficient than the last two.   If an inexpensive
source of molecular oxygen is available, it can provide a very efficient means
of oxidation.  The oxidation reaction consists of converting sodium sulfide
(Na2S) to sodium thiosulfate (Na2S203) according to the reaction:  2Na2S +
202 + H20 -> Na2S203 + 2NaOH.  The benefits derived from black liquor oxida-
tion are that it increases the black liquor solids and therefore, improves
the multiple-effect evaporation process, reduces odors, reduces the corrosion
rate of metal evaporating surfaces, reduces the chemical makeup requirements
for Na2S04 and CaO, and increases the sulfidity of the white and green liquor,
which affects pulp yield and quality.  It may, however, reduce the heating
value of the black liquor, and foaming may occur because of the presence of
fatty acid sodium salts, particularly when using softwoods.    The use of mo-
lecular oxygen in place of air, however, has proved to be effective in con-
trolling this problem.  The oxidation may be carried out as weak liquor (prior
to MEE) or as strong liquor after concentration.  Figure 3-36 shows a typical
strong liquor oxidation system. •
     The oxidation efficiency of the process is measured by the change in
sulfide  (Na2S) of the black liquor in grams/liter.  Conversion rates in.
excess of 99 percent are common.35'36  Acceptable sulfidity levels in the
liquor are on the order of 0.1 g/1 to achieve TRS emission limits from the
recovery boiler.  The efficiency is a function of residence time, liquor mix-
ing, liquor  height, air flow rate, and  inlet Na2S concentration.  The oxygen
requirement  for conversion is generally measured in cubic feet of 02/ton of
Na2S, pounds of 02/ton of pulp, or cubic feet of 02/gal of black liquor.
     Emissions of TRS generally come from the tank vent and are on the order
of  10 to 500 ppm.  The gas  volume  is between  16,000 and 48,000 ft /TADP.12
                                      .97

-------
                                FOAM
                               BREAKER
                                                           EXHAUST
                                                              AIR
AGITATOR
       AIR
             BLACK
             LIQUOR
             INLET
         BLACK
         LIQUOR
         EXIT
                         TURBINE
                         AERATOR
 Figure 3-35.   Agitated air sparging  system for black liquor oxidation.12
BLACK
LIQUOR
EXIT
           	L JL_I	
           NUMBER 2
           QXIQIZER
                                               STRONG BLACK
                                              LIQUOR STORAGE
                 BLACK
                 LIQUOR
                 INLET
                  BLOWER  PUMP
                                    BLOWER PUMP
                                                           »
     Figure 3-36.  Champion two stage unagitated strong  black  liquor
                           oxidation system.31*
                                   98

-------
3.2.2  Sources of Emissions and Control
       11 •— '    ' '-	...-••-•-•- 	    r,, * JT	 _.   ^,-
     Emissions from the pulping department consist primarily of odorous gases.
These gases are generally controlled by one of three methods:  1) reduction in
quantity through process operation,  2) scrubbing with fresh water or alkaline
solutions, or 3) incineration of noncondensable gases.  Table 3-5 summarizes
the control options that are available for each of the emission sources within
the pulping department.
     TABLE 3-5.  SUMMARY OF TRS CONTROL OPTIONS FOR PULPING DEPARTMENT
                                                                      12
              Emission source
  Control  option
     Digester gases  (blow and relief)
     Washer vents
     Evaporator gases

     Condensate stripping
     Condensate water

     Black liquor oxidation  (tower vent)
     Tall oil reactor vent
Incineration
Incineration3
Scrubbing
Incineration
Incineration
Steam stripping
Air stripping
Incineration
Scrubbing
 Diffusion and pressure washers.

 3.2.2.1   Digester and Blow Tanks--
      Digester and blow tank noncondensable gases are  generally  low  in  volu-
 metric flow  rates, but high in  such  TRS  compounds  as  methyl mercaptan,
 dimethyl  sulfide, and dimethyl  disulfide.  The  emissions  from the digester and
 blow tanks can range from  0.5 to  5.0 Ib  of sulfur  per ton of pulp  (10  to
 1000,000  ppm volume).
      Gases from  the relief system pass through  the turpentine recovery con-
 denser and decanting system,.and "the flow is  relatively constant at low volu-
 metric flow  rates.  When batch  digesters are  used, the emissions from  the blow
 tank are  periodic and can  occur up to 120 times per day (depending  on  number
 of digesters and the cooking time).   The time during  which peak flow occurs,
                                       99

-------
 however,  is only K) to .20 minutes per digester cycle.13   The  volume  of  gas
 released  depends on the digester volume,  the gas  temperature,  the moisture     !
 content,  the condition of hot water accumulator condensers, and  the  amount  of
 air and inert gases in the system.
      The  amount of gas generated also may be affected  by  upsets  in the  system.
 The normal  practice is to blow only one digester  at  a  time.   If  an upset occurs,
 a  simultaneous blow of two digesters will  generally  overload  the primary and
 secondary condensers and result in  an increase in noncondensable gas treat-
 ment volume.   Table 3-6 presents typical  volumes  during average, maximum, and
 upset  conditions.
                  13
              TABLE  3-6.  GAS  FLOW  RATES  FROM BATCH DIGESTER13
Condition
Average
Maximum
Upset
Digester blow
ft3/blow
1,600
4,000
125,000
ft3/TADP
128
320
10,250
     The amount of flow also varies between digester types, with the continuous
digester generating the lowest emissions.  Table 3-7 presents comparative data
                                                      12
for both batch relief and continuous digester systems.

   TABLE 3-7.  TYPICAL RANGES OF DIGESTER NONCONDENSABLE GAS FLOW RATES12
               Source
Flow rate ft3/TADP
             Batch blow
             Batch relief
             Continuous
 15,200 - 203,500
     10 -   3,040
     20 -     200
     Because of the nature of the digester TRS emissions, scrubbing is not an
effective control option.  Scrubbing is effective in removing acidic TRS species,
but it is not effective in removing other TRS species.  For effective
                                     100...

-------
destruction of all the gases, the noncondensable stream usually is incinerated
in either a lime kiln or a wood-waste-fired boiler.
     Because of the range of flow rates expected from the batch digester
blow, it is not economically feasible to design for peak flow conditions.  In
most systems a gas accumulator is used to collect the blow gases and retain
                                                37
them to provide uniform flow to the incinerator.
     The most common type of flow equalization system is the vaporsphere
(Figure 3-37).  The sphere contains a diaphragm and flow control system for
venting collected gases to the incinerator.  Vaporspheres are typically 10,000
ft3.17'37
     A second type of gas holder is a floating-cover system in which a water
seal is used  (Figure 3-38).    Either system must be maintained to keep
the gas concentration below or above the flammability limits.  Table 3-8
presents the acceptable flammability limits.38  Duct design and flash back
systems are used to prevent explosions.  Table 3-9 provides typical flame-
                     go
spreading velocities.
3.2.2.2   Washer Hood Vents-
     Existing hood vents over vacuum washers are not generally controlled
because of the low concentrations of the TRS in the gas stream, the large gas
volume, and the high amounts of water vapor.  TRS control is required of new
washers under NSPS.  The limit is 5 ppm by volume corrected to actual oxygen
content of the gas stream.  Emissions from pressure-type washers and diffusion
washers may be condensed to remove water and noncondensable gases and then
vented to the lime kiln or wood-waste-fired boiler to be incinerated.  Because
the volumeric flow rate of the gas stream  is constant, the use of a gas .equali-
zation system is not required.
3.2.2.3   Evaporator Condenser--
     Gasesxfrom multiple-effect evaporators (tail gas and hot well) are
generally condensed to remove water and condensable organics.  The remaining
gas may be scrubbed or incinerated in the  lime  kiln.  The volume of gas  re-
quiring treatment varies greatly !and depends on evaporator design, condenser
design, condenser temperature, and the amount of  infiltrated air.  Table 3-10
                                                                 12
presents a range of gas volumes for the two types  of condensers.
                                      101

-------
  Diaphragm
 Vacuum
Pressure
 Relief
      Relief   Gases
                              Sliding
                              Weight
                                     Flow
                                    Control
      Figure 3-37.  Vaporsphere flow equalization gas holders.17'37
         Blow
By-pass
 Vent
                                                   Vacuum
                                                  Pressure
                                                   Relief
                                                   System
                                                        Handling
                                                        System
   Figure 3-38.  Floating cover flow equalization gas holders.17'37


                                  102

-------
   TABLE 3-8.   FLAMMABILITY LIMITS IN  AIR FOR  KRAFT NONCONDENSABLE  GASES
                                                                        38

H2S
CH3SH
CH3SCH3
Terpene
Concentration, % by volume
Lower
4.3
3.9
2.2
0.8
Upper
45.0
21.8 ~
19.7
•*••
     TABLE 3-9.   FLAME-SPREADING VELOCITIES OF AIR-MERCAPTAN MIXTURES
                                                                     39
Mercaptan
% by vol.
18.9
22.8
23.1
23.7
25.5
25.7
Flame velocity,
ft/s
1.8
1.5
1.3
1.2
0.6
0.5
  TABLE 3-10.  TYPICAL RANGES OF EVAPORATOR NONCONDENSABLE GAS FLOW RATES
                                                                         12
Source
Surface condenser
Jet condenser
Flow rate, ft3/TADP
20 to 420
10 to 100
     The primary sulfur compounds that are emitted from the MEE are H2S and
CH3SH.  The concentrations range from 0.2 to 3.0 Ib sulfur per ton of pulp
(300 to 25,000 ppm volume).  The alkaline scrubber is effective in reducing
TRS compounds that are acidic (H2S and CH3SH), but it is not effective in re-
moving CH3SCH3 and CH3SSCH3.17
                                     103

-------
 3.2.2.4    Condensate Stripping--                                              :
     Noncondensable gases from the air stripping column or steam stripping    ''
 column, which are generally low in volume, are incinerated in the lime kiln or
 wood-waste-fired boilers.                                                     :
 3.2.2.5    Black Liquor Oxidation—
     The gas stream from the oxidation tower vent generally contains a low
 concentration of TRS and is saturated with water vapor.  This gas stream is
 generally  not controlled.  If control is necessary, however, a primary con-   i
 denser is  usually used to remove the water vapor.  The gas stream volume is
 generally  too large to be incinerated in lime kilns, but it can be vented to
 and incinerated in a wood-waste-fired boiler.
 3.2.2,6    TRS Scrubbers-
     Scrubbing a noncondensable gas stream to remove TRS compounds is generally
 accomplished with packed-bed scrubbers using a countercurrent flow of an      ;
 alkaline liquid (white water).  The packing medium can be gravel, stone, or
 packing rings.    The scrubbers can be constructed of mild steel or stainless
 steel (304 or 316L).  The stainless steels are used where corrosion is
 expected.                                                                     .
     Figure 3-39 shows a typical  two stage packed bed scrubber that is appli-
 cable to noncondensable gases from multiple-effect evaporators.  General
 liquor-to-gas ratios are on the order of 300 gal/1000 acfm of noncondensable
 gas.  The  superficial  gas velocity is approximately 50 ft/min through the
 packed section.  A mist eliminator is generally used to remove mist carryover.
 3.2.2.7  Incineration  Systems—
     A noncondensable  gas incineration system may consist of a flow equaliza-
tion system, condenser scrubber,  rupture discs, fan, flow recorder, condensate
traps, bypass vent system,  flame arrestors, and the incinerator.    The in-
cinerator may be the lime kiln, separate incinerator, or wood-waste-fired
boiler (Figure 3-40).
3.2.2.8  Lime Kiln Incineration—
     The noncondensable gases are introduced either into a dedicated burner or
into the primary air' system of the kiln burner for effective mixing with air
to ensure complete combustion.   Systems using a dedicated burner are generally
40
                                     104

-------
  SPRAY
 NOZZLES"
    SCRUBBED GASES

         A

           4-	_
               T
             600mm

             600
   •mm.
600mm
               t
             600mm
             600mm
             600mm
               1
                    /  \
                    ^.rSv^
                    \
      WHITE  LIQUOR
        RETURN
                    I.Q 50mm
                    (2 in.)
                                    I. D.  150mm (6 in.)
                       I.Q 600mm (24 in.)
                                    MIST ELIMINATOR
                       25mm (I in.) RASCHIG RING
                                               I.D. 150mm (6 in.)
                            HOTWELL  GASES
                            MAX. FLOW - lOOOm'/h
                                      (590ft5/min)
                               COOLED WHITE LIQUOR
                               MAX. FLOW - 50 mVh
                                         (30ft3/min)

Figure 3-39.   Hotwell gas  scrubber for 100 metric tons per hour 9400 gpm)
     evaporation plant for H2S-Separation of 95 percent  or more.40
                               105

-------
o
a>
                           Gases From Multiple-
                            Effect Evaporators
                           Gases  From Turpentine
                                  Condenser
                 Relief Gases
            from,
                  Kamyr Digester
                 Blow & Relief
                 Gases  From 4
                 Batch Digesters
Defibrator Off-
     Gases
                                         Heat
                                     Accumulator
                                      Entrapment
                                       Separator
                                                          Floating-Cover
                                                            Gas Holder
                                                      Vent
                                                                                           t
                                                                                        J
                               Figure 3-40.   Noncondensable gas incineration system.
                                                                                          1-7

-------
used on a lean gas collection system.  Rich gas streams are generally incinerated
in the kiln.  The dilution factor of the gases entering the kiln is about 50
to 1, and the gas velocity is at least 30 ft/s.12'41
     The gases are incinerated in the lime kiln at temperatures between 2200°
and 2550°F.  Residence times are generally greater than 0.5 s above 1000°F and
may exceed 1.5 s in longer kilns.  The S02 generated from the combustion is
adsorbed by the lime to form calcium sulfite (Ca2 SOg).
3.2.3  Malfunctions
     The following subsections discuss process and control equipment malfunc-
tions that have an impact on uncontrolled and controlled emission rates.
3.2.3.1  Digester Relief Systems--
     Most of the operating problems that occur with digester relief systems
have an impact on pulp quality, but some also have an impact on atmospheric
emissions.  Entrainment of air into the digester system as a result of improper
charging can produce higher than normal noncondensable gas volumes.  In some
mills excessive liquor carryover and foaming can cause plugging of the system
                                        2
and fouling of the turpentine condenser.   Foaming can occur as a result of
resins in softwoods and entrained air.  Table 3-11 summarizes the malfunc-
tions that can occur in the digester relief system.
                                                                             2,42
3.2.3.2  Digester Blow System—
     The primary malfunction that occurs in digester blow systems is the
fouling of the hot water accumulator condensers or pluggage of the blow tank.'
As the pluggage progresses the ability of the condenser to condense the flashed
steam is reduced.  This results in increased noncondensable gas volumes and
overloading of the vent control system and gas accumulator.  Serious malfunc-
tion of the condenser may cause vent gas system to be bypassed as a result of
top much pressure.  Figure 3-41 shows the effect of heat exchanger malfunc-
                                             17 42
tions on digester noncondensable gas volumes.  '
     Simultaneous blowing of two or more digesters can produce high noncon-
densable gas volumes and reduce the condenser's efficiency.  Because of this
increased volume, the vent system may be bypassed to the atmosphere.
     If the condensers become seriously fouled, the back pressure in the digester
blow system can cause fugitive emissions through duct flanges, access hatches,
                                     107

-------
        TABLE 3-11.  MALFUNCTIONS THAT MAY OCCUR IN DIGESTER RELIEF
                       TURPENTINE RECOVERY SYSTEMS2'1*2
   Malfunction
     Primary effect
                                                            Result
Liquor carryover
Low water flow rate to
turpentine condenser


Failure to close  di-
gester valve after blow
Digester relief line plug-
gage
Trapped air in chips due
to poor digester filling

Liquor foaming when cooking
resinous woods
               f-

Temperature condenser plug-
gage or fouling
Increased condenser water
temperature
Fouling of digester blow
line
Digester overpressure
and emergency bypass
relief or premature
digester blow which
will increase TRS
emissions
Reduced condenser heat
transfer and increased
noncondensable gas
temperature yielding
increased equilibrium;
TRS vapor pressure and
TRS emissions

Increased equilibrium
TRS vapor pressure and
TRS emissions

Loss of digester cook-
ing liquor, overpres-
sure during blowing,
increased digester blow
volume which will in-
crease TRS emissions
                                      108

-------
I
    20


    io H


     o
             23
             j
                 V( t)  dt = 550m3/blow
           MEAN - 1400 m3/h
                                                  25
                5        10       15       20
                       TIME, MINUTES

                CASE  I.    NORMAL OPERATIONS
   40  -


 ^ 30  -
 *>.
V
WQ 20  -


 1  »H
            23
            )5   V( t ) dt = 300m3/blow
                                          20
                                                   25
       0        5        10        15
                       TIME, MINUTES

CASE 2.    MALFUNCTION OF BLOW HEAT RECOVERY HEAT EXCHANGERS
    50 -



    40 -



    30 -


    20 -



    10 -



    0
             23
             5 V(t) dt = 4500m3/blow
             \^
          MEAN-ll,700m3/h
               5
                                          20
                                                   25
                    10        15
                  TIME, MINUTES

CASE 3.     INCREASING MALFUNCTION  OF HEAT  EXCHANGERS
Figure  3-41.
              Kraft batch dige'ster blow gas flow after condensing
                  and without equalization.17
                            109

-------
etc.  Bypass of the primary condenser has been observed as a result of severe

corrosion of the blow tank and associated ductwork.   It should be noted that

the concentration of TRS in these streams is very high, and the inspector should

wear appropriate respirators in these areas when fugitive losses are noted.
     The blow tanks are pressure-rated vessels and can be weakened by corrosion

and repeated overpressure.  Table 3-12 summarizes the malfunctions that may

occur in the digester blow system.
         TABLE 3-12.  MALFUNCTIONS THAT MAY OCCUR IN DIGESTER BLOW
                  TANK HOT WATER ACCUMULATOR SYSTEMS2*17>*2
   Malfunction
     Primary effect
     Result
Fouling of primary hot
water accumulator, pri-
mary and secondary con-
densers

Low water flow rate to
primary and secondary
condensers or hot water
accumulator

Simultaneous digester
blow
Corrosion of blow tank
and ductwork
Reduced heat transfer and
loss of condensate
Increased condenser water
temperature
Overloading of primary and
secondary condenser and/or
TRS incineration system
Fugitive condensable gases
Increased digester blow
volume and increased
TRS emissions
Increased equilibrium
TRS vapor pressure and
TRS emission rates
Increased noncondensable
gas volume, carryover
of condensate into vent
gas system, overpres-
sure bypass to at-
mosphere, loss of cook-
ing chemicals and in-
creased TRS emissions

Increased TRS emissions
as a result of vent
system by pass
3.2.3.3  Multiple-Effect Evaporators—
     The primary malfunctions that affect evaporators are associated with foul-
                                                        2 12 21
ing of the evaporator body and ambient air infiltration. '  '    Both can

result in reduced evaporator efficiency.  Ambient air infiltration results

in increased noncondensable gas volumes and reduced condenser performance,

which can result in low water temperature and increased equilibrium TRS vapor

pressures.43   In systems that use alkaline scrubber noncondensable gas control,
                                      110

-------
loss of water flow-, reduction in alkali content, or an increase in gas volume
can result in lower mass transfer rates and increased TRS emission rates.
Table 3-13 summarizes the malfunctions that may occur in multiple-effect
                   ? i? /n M
evaporator systems. §1*HfH
3.2.3.4  Black Liquor Oxidation-
     Malfunctions in the black liquor oxidation system generally result in
reduced oxidation efficiency (high outlet sulfidity) which produces elevated
TRS concentrations from the recovery boiler.  Malfunctions include reduced
air flow volume, pluggage of oxygen injection system, increased black liquor
flow rates, liquor foaming, and increased inlet liquor sulfidity.   Table 3-14
                                           33
summarizes the malfunctions that may occur.
3.2.3.5  TRS Vent Control System--
     The TRS vent control system collects noncondensable gases from the blow
tank, hot water accumulator, and other miscellaneous systems.   Increases in
gas volume may exceed system design and limit the ability of the sources to
be evacuated.  Increased flow rates result in higher static pressure losses
through the lines and may require larger gas-moving equipment.
     For safety reasons, the concentration of the organic vapor must be main-
tained either below the lower explosive limits~or above the upper explosive
limits.  Flash-back flame protection devices must be used to prevent explosions.
     Entrained moisture must be removed from the gas stream to prevent cooling
or blowout of the incinerator flame.  Large variations in flow must be mini-
mized to stabilize flame temperature and maintain acceptable residence time.'
Table 3-15 summarizes the malfunctions associated with the noncondensable
gas incineration system.
3.2.4  Inspection of Pulping Department
     Because emissions from the pulping department are gaseous in nature and
the gases are potentially hazardous at high concentrations, physical measure-
ment and evaluation of emission sources are limited.
     Inspections of these sources are generally based on information concerning
equipment specifications, process procedures, process weights, and/or control
equipment bypass and malfunction.  The ability of the inspector to measure
TRS concentrations with hand-held equipment is extremely limited.  Errors in
45
                                     111

-------
        TABLE 3-13.  MALFUNCTIONS THAT MAY OCCUR IN MULTIPLE-EFFECT
        	EVAPORATOR SYSTEMS2»12»"3»*"
   Malfunction
     Primary effect
     Result
Evaporator fouling and
scaling
Ambient air inleakage
into evaporator body
Low condenser water
flow rate
High inlet condenser
water temperature
Reduced scrubber water
flow rate
Increased scrubber gas
volume
Scrubber packing flow
channeling
Liquor foaming
Separation of soap
from liquor
Carbonaceous deposits
in evaporator body
Reduced evaporator effi-
ciency


Increased noncondensable
gas volume
Increased condenser outlet
water temperature
Increased condenser outlet
water temperature
Reduced liquor-to-gas ratio,
reduced mass transfer and
adsorption

Increased superficial veloc-
ity through scrubber, re-
duced liquor-to-gas ratio,
reduced mass transfer and
adsorption, increased mist
eliminator carryover

Reduced liquor-to-gas con-
tact, reduced mass transfer
and adsorption

Liquor carryover and re-
duced heat transfer
Liquor carryover and re-
duced heat transfer
Liquor carryover and re-
duced heat transfer
Reduced boiler effi- .
ciency and increased
TRS emissions

Reduced condenser
efficiency and in-
creased TRS emissions

Increased condenser
equilibrum TRS vapor
pressure and TRS emis-
sions

Increased condenser
equilibrium TRS vapor
pressure and TRS emis-
sions

Increased TRS emissions
and decreased removal
efficiency

Increased TRS emissions
and decreased removal
efficiency
Increased TRS emissions
and reduced removal
efficiency

Reduced evaporator
efficiency, decreased
final liquor solids
attainable and in-
creased TRS emissions

Reduced evaporator
efficiency and increased
TRS emissions    .    '

Reduced evaporator
efficiency and in-
creased TRS emissions
                                     112

-------
     TABLE 3-14.  MALFUNCTIONS THAT MAY OCCUR IN BLACK LIQUOR OXIDATION
                                   SYSTEMS
   Malfunction
     Primary effect
     Result
Reduced air flow volume
through oxidation tank
Plugging of air sparge
Increased liquor flow
rate
Liquor foaming
Increased inlet liquor
sulfidity
Reduced oxidation of sodium
sulfide
Stratification of liquor air
column and reduced contact
Decreased liquor residence
time and oxygen adsorption
Foam carryover limits
system liquor volume and
blowing rates
Increased outlet liquor
sulfidity and TRS emis-
sions

Increased outlet liquor
sulfidity and TRS emis-
sions

Increased outlet liquor
sulfidity and TRS emis-
sions

Increased outlet liquor
sulfidity and TRS emis-
sions

Increased outlet liquor
sulfidity and TRS emis-
sions
      TABLE 3-15.  MALFUNCTIONS THAT MAY OCCUR IN NONCONDENSABLE GAS
                             INCINERATION SYSTEM
   Malfunction
     Primary effect
     Result
Increased flow volume
Operation between lower
and upper explosive
limits

Low gas flow velocity
Entrained moisture
Exceeding system design,
reduced residence time
Operation below flame
propagation velocity

Flame blowout, reduced
flame temperature
Fugitive TRS emissions,
increased TRS emissions
due to incomplete com-
bustion

Potential for explosion
Potential for explo-
sion and/or fire

Increased TRS emissions
as a result of incom-
plete combustion and
potential for explo-
sion
                                     113

-------
measurement due to interference are a major concern.  In addition, some sources
are vented through small pipes at high velocity, which makes it extremely
difficult to measure flow.
     Because the emissions are gaseous, visible emission evaluations are not
useful except for fugitive emissions in the area between the blow tank and
hot water accumulator.  In this area the presence of a leak is observed as a
condensed water plume.
     Special care must be taken in such areas to avoid confined spaces in which
toxic concentrations of hydrogen sulfide may accumulate.  The inspector should
carefully avoid any contact with process vent plumes that may contain TRS     ;
gases.
     The following discussions concern information that should be documented
for each process step to allow calculation of process weights, operating cycles,
and control equipment performance.  In many instances, the data required do
not apply to the calculation of emission rates from a specific source, but are
necessary for analysis of other process operations in the chemical recovery
cycle.
3.2.4.1  Digester—
     The inspector should determine the cooking cycle of the digester, the
charge weight, the white liquor charge weight, the liquor sulfidity, and the
digester volume.  The number of cooks per day is an indication of the level   :
of operation, assuming reasonably constant pulp yield.
     The volume of the blow tank and blow time, together with the calculated
blow volume, can be used to calculate the size of primary and secondary con-  '
densers in the hot water accumulator.  The occurrence of simultaneous digester
blows can overload the system.  The inspector should observe the blowing opera-
tion to determine if condensable gases are being lost from ducts as a result
of overpressure in the system.  These fugitives bypass the TRS nbncondensable
gas system.
     Malfunctions concerning the blow tank system occur primarily as a result
of condenser plugging.  The final  equilibrium vapor pressure of TRS gas in the
gas stream is defined by the secondary condenser temperature and gas stream
pressure.  In cases of condenser plugging, the heat exchange rate across the
condenser and the outlet water temperature are reduced.   The inspector should
                                     114

-------
review operating charts for the condenser water temperature differential  (inlet/
outlet) to determine if a trend is occurring that would indicate fouling.
     In some systems the noncondensable gas volume is monitored and recorded
by automatic instruments.  The inspector should review these data to~determine
if peak gas flow is within the design specifications of the incineration  system.
It is suggested that the inspector request and maintain a process flow diagram
for the noncondensable gas system.  Table 3-16 provides an inspection checklist
for the digester system.
3.2.4.2  Digester Relief—
     The digester relief system contains significant quantities of TRS compounds
and turpentine.  The amount of turpentine is a function of cooking variables,
wood species, and time of year.  The resinous woods produce the highest quantity
of turpentine.  To evaluate the Operation of the relief system, it is necessary
to know the relief gas flow volume and cycle time.  The peak loading and  highest
temperature represent the most critical period.  The digester relief gases are
condensed in a surface condenser.  The inspector should determine the final
outlet gas temperature of the condenser (inlet/outlet), and the vacuum if steam
ejectors are used.  Changes in outlet gas temperature can indicate higher
emission levels as the vapor pressure increases.  Table 3-17 is an inspection
checklist for the digester relief system.
3.2.4.3  Brown Stock Washers--
     The inspection of brown stock washers consists of determining source loca-
tion, process weight, and gas volumes.  Emission levels are generally low and
not controlled.  Table 3-18 is an inspection checklist for the brown stock
washer system.
3.2.4.4  Multiple-Effect Evaporators--
     The inspection of multiple-effect evaporators consists of documenting the
emission points, process weights, and control equipment used.  Where condensers
are used, the inspector should record the cooling water flow rate, condenser
surface area, gas stream pressure (where ejectors are used), and the differ-
ential water temperature (inlet/outlet).  Table 3-19 is an inspection check-
list for multiple-effect evaporators.
                                      115

-------
       TABLE 3-16.  CHECKLIST FOR INSPECTION OF DIGESTER BLOW SYSTEMS

     Digesters
     Number of digesters  	
     Chip charge weight   	 tons
     White liquor charge weight
ft°/batch
     Digester volume   	ft
     Cook time        	min
     Blow time        	min
     Cooks per day    	
     Blow Tank
                                3
     Blow tank volume  	ft
     Pre-blow digester temp.    	 °F
     Final blow tank temp.      	°F
     Blow time      	min
     Calculated blow volume   	lb/hc
     Hot Water Accumulator
     Hot water accumulator condensers
                    	 Primary
                    	 Secondary
     Primary condenser temperature  _
                    Water flow
Type
Type
   o
    F
   gpm
     Secondary condenser temperature
                    Water flow
                    Surface area
D    TRS Control
     gpm
     ft2
     Noncondensable gas volume
                         Temp.
                         Pressure
 acfm
  in H20 (vacuum)
(continued)
                                     116

-------
TABLE 3-16 (continued)
     Vapor equalizer     Type
                         Volume
fr
     Noncondensable gas treatment method
               	 Incineration
               	 Scrubbing
     Incinerator type
                      Catalyst
                      Direct fire
                      Lime kiln
                      Wood-waste boiler
                      Fossil fuel boiler
                      Recovery boiler
     Dilution ratio at burner 	
     Explosion protection     	yes
     Type:     	Flame arresters
               	 Bypass vent
               	 Rupture discs
             no
                                      117

-------
 TABLE 3-17.  CHECKLIST FOR INSPECTION OF DIGESTER RELIEF SYSTEMS
Number of digesters 	
Relief flow rate 	
  acfm
Noncondensable gas flow
Condenser type
  Water flow
         acfm
         Or-
gpm
  Inlet temperature _
  Outlet temperature
  Surface area
ft'
  Operating pressure
Turpentine recovery _
Noncondensable gas treatment
      in. H20 (vacuum)
     gal/TADP
     gph
              scrubber
              incinerator
              none
                                118

-------
     TABLE 3-18..  CHECKLIST FOR INSPECTION OF BROWN STOCK WASHER SYSTEMS
Number of wash lines
Number of stages per line
Washer type vacuum	
     pressure
     batch
diffusion _
continuous
Process weight
TADP/day
Outlet weak black liquor flow
               M Ib/h
               gpm
Outlet weak black liquor solids
Exhaust hood gas flow rate 	
Exhaust hood gas temperature 	
            acfm
            0,-
Control method
incineration
condenser
none
Noncondensable gas flow rate
Condenser water flow rate
Condenser water inlet temp
Condenser water outlet temp
Condenser surface area
              acfm
              gpm
              Or
              ft'
                                      119

-------
 TABLE 3-19.  CHECKLIST FOR INSPECTION OF MULTIPLE-EFFECT EVAPORATOR  SYSTEMS
A.
Multiple-Effect Evaporators
Number of evaporator lines _
Number of effects
Manufacturer
Inlet conditions
  Black liquor flow 	
       Black liquor solids
     Outlet conditions
       Black liquor flow _
       Black liquor solids
     Soap recovery       _
     Control type:  Tail gas
                    Hot well
     Noncondensable gas control
       Scrubber 	
       Incineration 	
       None            	
Weak black liquor oxidation 	
B.   Noncondensable gas scrubber
     Type	
     Superficial velocity
     Liquor to gas ratio
     Liquor type
     Liquor flow rate
     Pressure drop
                                    gpm
                                    M Ib/h
                               gpm
                               M lb/he
                               %
                               Ib/h
                              yes
no
                                ft/min
                                gal/1000 acfm
                                water
                                white liquor
                                alkaline solution
                                gpm     •     •
                                in. H20
 (continued)
                                     120

-------
TABLE 3-19 (continued)
C.
     Outlet gas temperature
     Liquor alkalinity
            PH
Tail Gas Condenser/Hot Well Condenser
Number of stages 	
Condenser type	barometric
     Condenser area
       Pressure
       Water flow
       Inlet temp.
      ' Outlet temp.
      Condenser  inlet flow
                Temperature
      Noncondenser  gas  flow
                          surface
                          ft2
                          in. H20 (vacuum)
                          gpm
                                acfm

                                °F

                                acfm
                                Or-
                                      121

-------
 3.2.4.5  Black Liquor Oxidation—
      Because the black liquor oxidation system determines the emission of TRS
 gases from direct-contact recovery boiler systems, the evaluation of the system
 is directed toward the final liquor sulfidity.  The inspection is directed
 toward determining the operating parameters that affect the oxidation effic- :
 iency of the system.  Table 3-20 is an inspection checklist for black liquor
 oxidation systems.
 3.2.4.6  Condensate Stripping—
      The inspection of the stripping system consists of measuring such key
 operating variables as condensate pH,  steam flow, air flow,  and condensate
 flow rate.   Table 3-21 is an inspection checklist for stripping systems.

 3.3  CHEMICAL RECOVERY
      The two major sources within the  chemical  recovery portion of the kraft
 pulp mill are the kraft  recovery furnace or boiler and the smelt dissolving
 tank.  The  kraft  recovery boiler is  used to combust spent liquor from the
 pulping  process.   The smelt dissolving tank,  located  below the  recovery boiler,,
 is  a large  vessel  that continuously  receives  a  molten  mixture of sodium sul-
 fide and  sodium carbonate from  the floor of the recovery  boiler.
 3.3.1  Recovery Boiler
     The  following  subsections  describe the  kraft recovery boiler process,
 identify  the  sources  of emissions from a  recovery boiler,  discuss the  con-
 trol techniques used  to minimize  emissions  from the recovery boiler,  discuss
 the  possible malfunctions associated with operation of the recovery boiler,
 and  present inspection procedures for  the recovery boiler  and associated con-
 trol equipment.
     The particulate matter and TRS emission rates from the recovery boiler
 depend on several interrelated boiler  operating variables; e.g.,  firing rate,
 black liquor heat valve, black liquor  concentration, total combustion air,
 primary air, black liquor sulfur-to-sodium ratio, primary  air temperature, char
 bed temperature, and black .liquor chlorine content.  Section 3.3.1.2 provides,
a detailed discussion of how these variables affect the uncontrolled emission
rate of the boiler and the overall performance of the electrostatic precipita-
tor (ESP) used to control  the particulate emissions from a recovery boiler.
                                     122

-------
   TABLE 3-20.   CHECKLIST FOR INSPECTION OF BLACK LIQUOR OXIDATION  SYSTEMS
A.
Black Liquor Oxidation
Oxidation type	weak
                        Air
                                        strong
     Oxidation method
     Weak black liquor
       oxidation (BLOX)
                             Oxygen
                          Porous plate diffusers
                          Sieve tray tower
                          Packed tower
                          Agitated air sparging
                          Rotating fluid contactor
                          Other
     Strong BLOX
     Liquor flow rate
                   Single-stage unagitated air sparging
                   Two-stage unagitated air sparging
                   Agitated air sparging
                  	gpm
                        M Ib/h
     Liquor solids content _
     Residence-'time	h
     Inlet liquor sulfidity _
     Outlet liquor sulfidity
     Oxidation efficiency 	
     Air flow volume
                             _g/liter
                             _ g/liter
                       acfm
     Tower  vent TRS control
     Control  type 	
                              yes
no
                    Condenser
                    Scrubber
                    Incineration
 (continued)
                                      123

-------
TABLE 3-20 (continued)
                           gpm
                           0,-
Condenser type _
  Surface area _
  Water flow
  Inlet temp
  Outlet temp  	°F
  Operating pressure 	
Noncondensable gas volume
              Temperature
Scrubber type 	
Superficial velocity 	
Liquor/gas ratio
Liquor type 	
                                 in. H20 (vacuum)
                                      acfm
                                 ft/min
                             gal/1000 acfm
     Liquor flow rate
     Pressure drop 	
Water
White liquor
Alkaline solution
	gpm
                     in. H20
     Outlet gas temperature
     Liquor alkalinity 	
                    PH	
     Incinerator type
                              Catalyst
                              Direct fire
                              Lime kiln
                              Wood-waste boiler
                                     124

-------
TABLE 3-21.  CHECKLIST FOR INSPECTION OF CONDENSATE  STRIPPING  SYSTEMS
 Condensate Stripping Systems
 Type	  Air	
           Steam
 Foul condensate flow rate
 Air Stripping
 Column type:  Tray
 Number of trays 	
 Tray distance 	
 Column diameter 	
 Air volume
              M  Ib/h
              gpm
      Other
 in.
     in.
 Condensate inlet temperature
 Condensate pH ________
 Steam stripping
 Column type:  Tray
 Number of trays 	
 Tray distance 	
 Column diameter 	
 Steam flow
       Other
   in.
     in.
Ib/h
 Condensate inlet temperature
 Condensate pH     .	
                                 125

-------
 TRS and S02 emissions are controlled by the optimization of process  variables
 that cause sulfurous gases to become chemically combined with sodium to  form
 a particulate emission which is collected by the ESP.
      The malfunctions that the increase emissions from the recovery  boiler  are
 generally divided into two areas:   1) those that occur as a result of furnace
 operation,  and 2) those that occur as a part of control  equipment operation.
 Several  malfunctions that result from improper boiler  operating  practices also
 have an  impact on ESP performance.   Section 3.3.1.4  provides  a detailed  dis-
 cussion  of both boiler and ESP malfunctions that affect  emissions.   A summary
 table (Table 3-22 on page 184) is  provided  that lists  the key parameters and   :
 potential malfunctions and their affect on  the overall emission  rates  associated
 with recovery boiler operation.
      A considerable  amount of information is  presented in Section 3.3.1.6 on
 the inspection of recovery boilers  and  ESP's.   The specific kinds of data that
 should be collected  (Table 3-23  on  page 191)  during  the  inspection is  identified
 along with  a  discussion  of the procedures and  calculations  that  should be used
 to  evaluate the inspection data  to  assess the  overall continuous compliance
 status of the recovery boiler operation.
 3.3.1.1  Process  Description--
      The kraft recovery  boiler or furnace is an  indirect  water-walled
 steam generator used  to  produce  steam and to recover inorganic chemicals
 from  spent  cooking liquors.   The boiler  consists of a large vertical  combus-
 tion  chamber  lined with water  tubes.. The heat exchanger  section typically
 consists of a  low-pressure  boiler,  superheater, and economizer.  Figure 3-42
 shows  a cross  section  of a modern Babcock and Wilcox (B&W) boiler.12  The fuel
 used  in the boiler is  spent concentrated cooking liquor  (black liquor).  The
 liquor in the  burners  has a solids  content of between 60 and 70 percent (depend-
 ing on wood species and yield) and  is made up of organic and inorganic frac-
      47 40
tions.  '    The organic fraction contains lignin derivatives, carbohydrates,
                2 49
soap, and waxes.  »    The inorganic portion consists primarily of sodium
sulfate.
     Black liquor is sprayed into the furnace at an elevated level  in the com-
bustion chamber through a number of steam or mechanical atomizing nozzles.   The
suspended liquor is burned as it falls through the combustion zone.   The follow-
ing are the major steps in the combustion process:46

                                     126

-------
                                               Furnace
                                               Slag Screen
                                               Tertiary Air Ports
                                               & Windbox
          Black Liquor
      Oscillator Burner \
                                        /Oscillator Burner
                                               Secondary Air Ports
                                             /& Windbox
   Pin Stud
Upper Limit
      Smelt Spouts
           &Hood
                                               Primary Air Ports
                                               & Windbox
  Green-Liquor
  Recirculation
    Pumps
             Dissolvingj|j Dissolving Tank
Figure  3-42.    Cross section of B&W recovery boiler.   '
                                  127

-------
      o    The liquor is dehydrated to form a char.
      o    The char is burned in a bed at the bottom of the furnace.
      o    The ash (inorganic portion) remaining after combustion  of  the  char
           is exposed to active reducing conditions  to convert sodium sulfate ;
           and other sodium-sulfur-oxygen compounds  to sodium sulfide.        ;
      o    The organic materials in the upper section of the furnace  are
           oxidized to complete combustion.
 Reduced inorganic material  (smelt),  which consists  of a mixture of sodium  sul-
 fide and sodium carbonate,  is continuously drained  from the furnace.  The
 ratio of sodium sulfide to  sodium carbonate depends on the temperature and
 the  ratio of sulfur to sodium in  the fired liquor.
      Combustion of the char begins on the hearth of the furnace.  Air for
 combustion is supplied through air ports  located in the furnace walls.   The
 primary air supply is  used  to initiate char combustion.  The primary air
 supply,  which is  introduced in the lower  portion of the char,  is  kept to a
 minimum to maintain  the necessary reducing  conditions  to convert  the ash to
 sodium sulfide.   The secondary air supply is  located  at a  higher  level in the
 furnace  to create the  oxidizing condition necessary to control the char  bed
 height.   A tertiary  air supply may be used  to complete combustion in the upper
 levels of the furnace  and thereby eliminate reduced sulfur compounds.  As the;
 char  bed  is  burned,  the inorganic ash is  liquified  and drained to the furnace
 hearth, where it  is  reduced.
      Figure  3-43  shows  the  location  of air  ports for  the two major American-
manufacturered  recovery boilers.   One uses  primary, secondary, and tertiary
air in finite  zones  or  levels,  and the other uses primary  air with secondary
                             12
air introduced  tangentially.    Combustion  gases produced  by the burning of
the liquor are  passed through  the  heat exchanger section of  the boiler before
being exhausted to a particulate  control  device.  The  gases are cooled to
about 805°F  in  the boiler tube  bank  before  passing  into  the economizer.
Temperatures of the gas  leaving the  economizer, which  are about 750°F, are
reduced in either indirect-contact or direct-contact evaporators.
     There are three types of direct  contact evaporators:  cyclone,  venturi,
and cascade.  Figures 3-44,  3-45, and 3-46 show the basic design for each.
Cyclone evaporators concentrate the black liquor by placing it in  contact
                                      128

-------
  BABCOCK 8 WILCOX
           Steam
                           12
COMBUSTION ENGINEERING
          fSteam
  LEGEND
  I. Furnace
  2. Smelt  Spouts
  3. Black  Liquor Spray Nozzles
  4. Primary  Air Supply
  5. Secondary  Air Supply
  6. Tertiary Air Supply           •
  7. Position  of Char Bed Burners  for Oil  or Gas
  8. Normal  Configuration of Char Bed
  8*. Same at  Low  Primary Air  Flow and  Pressure
  9. Screen  Tubes
  10. Superheater
  II. Boiler Tube Bank
  12. Exit  to Economizer
                       SECTION A-A
Figure  3-43.  Difference  in  air systems in U.S.  recovery boiler designs.
                                                                               12
                                         129

-------
                                 WALL-WETTING
                                   NOZZLES
                                   CYCLONE
                                  EVAPORATOR
  BLACK-UQUOR
  RECIRCULATINS
     PUMPS
MECHANICAL
  POWER
  STRAINER.
UOUOR TRANSFER
 TO SALT CAKE
   MIX TANK

ILACK-UOUOR
 INLET SPRAYS '
  FLUE
  GASES
Figure 3-44.   Cyclone evaporator.
                                           48
Figure  3-45.   Cascade evaporator.

                     130
                                          12

-------
                       FLUE GAS
                      FROM BOILERrA
CLEAN GAS OUTLET TO FAN
             WALL WASH
RECYCLE LIQUOR
60-70% SOLIDS
   190°F
LIQUOR TO BOILER
60-70% SOLIDS
                                                            YCLONIC SEPARATOR
                                                             MAKE-UP
                                                              WATER
                                                         LIQUOR FROM
                                                        MULTIPLE-EFFECT
                                                         EVAPORATOR
                                                        40-50% SOLIDS
                     Figure 3-46.  Venturi evaporator.
                                                      48
                                     131

-------
 with the high-temperature gas stream by using the wetted wall of the cyclone.48
 The cyclone removes approximately 50 percent of the uncontrolled particulate
 from the gas stream.  A venturi evaporator concentrates black liquor by placing
 the flue gas in contact with liquor through the generation of liquor droplets.
 The droplets are generated through the shearing action of the gas stream as
 it passes a weir into which the liquor is being pumped.  Venturi evaporators
 remove approximately 85 percent of the uncontrolled particulate emissions
 generated by the furnace when operated at 4 to 5 in.  H20 pressure drop.   In
 the cascade evaporator, a thin film of liquor coats several  tubes rotated
 through the flue gas stream.   This type of evaporator will  generally increase  ;
 the black liquor soTids content from 48 to 65 percent.  The rate of evapora-
 tion is related to the flue gas temperature and the cascade rotation rate.
 The evaporator can be operated beyond design rates  without  substantial  process
 upsets, and can remove up to  50 percent of uncontrolled particulate emissions
 from the furnace.
      As a result of the direct contact of flue gases  with the black liquor  in
 the evaporator,  there is  considerable stripping of  TRS compounds (i.e.,  generally
 H2S)  from the  liquor.   The  loss  of TRS  compounds is primarily the result of a
 reaction between carbon dioxide  (C02)  and  sodium sulfide in  the  liquor.   Gases
 from  the recovery  boiler  usually contain  12  to  15 percent C02.   When  the liquor
 is  exposed  to  the  gas  stream,  it absorbs  the  C02 and  the liquor  pH  is de-
 creased.  The  amount  of H2S released  from  the  reaction (Na2S  + C02  +  H?0 ->
 Na2 C03 + H2S)  increases as the  pH is decreased  and the  concentration of sodium
 sulfide is  increased.   One method of  reducing sodium sulfide  concentration  is
 to convert  it to sodium thiosulfate through the,use of black  liquor oxidation.
     Sulfur dioxide in  the flue  gas stream also  can result in the release of
 HgS as  a result of decreasing pH.  In many operations, however,  the level of
 S02 is  too low to have  any significant effect on the H2S emissions.
     The amount of oxygen in the flue gas stream has an  effect on the genera-
 tion of TRS gases from the combustion process but appears to have negligible
 effect on the stripping of H2S from the evaporator.                              :
     There are three types of indirect contact evaporators.   Figure 3-47 shows
the basic concept employed by each.1  In general, these evaporators evaporate
the liquor by use of a noncontact tube and shell design.  Because the black
                                     132

-------
    B & W HIGH SOLIDS SYSTEM WITH NO DIRECT CONTACT EVAPORATOR
                  ADDED
Ur-
i
BLACK l
LIQUOR
~l




•i
PRIMARY
PARTICULATE
CONTROL
DEVICE
•



SECONDARY
PARTICULATE
CONTROL
DEVICE
(OPTIONAL)
                       CIRCULATION
                       EVAPORATOR
    CE SYSTEM WITH NO FLUE GAS DIRECT CONTACT EVAPORATOR
          s I           CONTACT  ,
          ^L«-  -»JEVAPORATOR |___

               !  I BLACK LIQUOR
               I  •« •» mm MB ^ •• *
    CE SYSTEM WITH NO DIRECT CONTACT EVAPORATOR
        ^*—JrF
               BLACK LIQUOR
                          FORCED   1
                        CIRCULATION '-
                        EVAPORATOR
Figure 3-47.  Three types  of indirect contact evaporators.
                               133

-------
liquor in a noncontact evaporator design does not come in contact with the  ,
flue gas, stripping of TRS compounds is prevented.  Evaporated water is con-
densed by using a tail gas condenser.  Noncondensable gases are directed to
lime kilns or into the furnace primary air system for incineration.
     The high temperatures in the furnace char zone result in a partial vapori-
zation of sodium and sulfur from the smelt.  The fume is removed from the
furnace with the combustion gases and condenses to a fine particulate consist-
ing of sodium sulfate (Na2S04) and sodium carbonate (Na2C03).
     Modern recovery boilers are sized for two process conditions:  1) the
heat input to the furnace, and 2) the weight of the chemicals to be recovered.
Both of these conditions affect the heat release rate as a function of
                      '3                                                  2
furnace volume (Btu/ft ) and furnace cross section or hearth area  (Btu/ft ).
Typical design values are on the order of 9800 Btu/ft  (furnace volume) and
              o               Cf\
900,000 Btu/ft  (hearth area).    The exact dimensions of the furnace depend
on the elemental composition, solids content, heat value, sulfidity, and
chloride content of the black liquor.  Deviations of 10 percent or greater in
                                                                     50
design variables should be investigated to ensure maximum efficiency.    The
manufacturers generally consider the boiler to be overloaded when the
firing rate (Btu/h) exceeds 120 percent of the rated value.
     Operation outside of design values can increase emission rates as well
as reduce thermal efficiency as result of tube fouling and reduced smelt
recovery.
3.3.1.2  Sources of Emissions-- •
     The uncontrolled partfculate matter and TRS emission rates from the
boiler depend on several interrelated boiler operating variables; e.g., .firing
rate [pounds of black liquor solids (BLS) per hour], black liquor heat value
(Btu/pound BLS), black liquor concentration (% solids), total combustion air
(excess air), primary air (% of total air), black liquor sulfur-to-sodium
ratio (S/Na2), primary air temperature (°F), char bed temperature  (°F), and
black liquor chlorine content (55).51»52,53,54  In add1tion to affect1ng the
uncontrolled emission rates, these variables can also reduce ESP performance.
The following discussion addresses both of these effects.
     It should be noted that reductions in TRS and S02 emissions from kraft
furnaces result primarily from the optimization of process variables that
cause sulfurous gases to become chemically combined with sodium to form a

                                     134

-------
particulate emission.  Operation of the boiler under these process conditions
can also result in increased uncontrolled particulate emissions in the form of
sodium sulfate.
     Firing Rate—The firing rate of a kraft recovery boiler is measured in
pounds of black liquor solids per unit of time (either pounds BLS/24 hours or
pounds BLS/hour).   Given a specific heat value of the black liquor solids,
percent solids in the liquor, and elemental  composition of the liquor, the
flue gas volume produced and boiler heat input can be defined.  The firing
rate of a recovery furnace is often increased to increase pulp production
rates.  Increasing the firing rates, usually requires an increase in the
gallons per minute of the fired liquor.  The firing rate is limited by the
pumping capacity of the system based on the  liquor temperature and viscosity.
     The-boiler combustion chamber is designed (i.e., sized) according to the
expected heat release rate and volume of flue gas at maximum firing condi-
tions.54  An increase in the flue gas volume above design conditions causes an
increase in vertical  gas velocity through the furnace combustion zone and an
increase in particulate emissions resulting  from the entrainment of black
liquor droplets and char particles.  Thus, particulate emissions tend to in-
crease with increased velocity.  The loss of sodium-based particulate may
increase as the flue gas volume is increased because of the higher gas
velocities and temperatures on the hearth of the furnace.  The release of
sodium in the flue gas from the char bed increases as flue gas volume increases
because of favorable diffusion conditions.55  The release of sodium is also
related to the temperature of the bed and is primarily the result of evapora-
tion.  The rate of evaporation depends on the diffusion conditions (gas
                                                            CfT C~J             '
velocity) in the zone between the flue gas and the char bed.  V           ,
     An increase in flue gas volume also increases the velocity passing
through the steam tubes and decreases the efficiency of heat transfer.  This
decrease in thermal efficiency reduces the overall boiler efficiency and in-
creases the stack temperature.  Because of the increased gas volumes and their
effect on char bed temperature, TRS emissions also tend to increase with
boiler firing rate.  Figure 3-48 shows the effect of firing rate on TRS and
steam generation rates for a typical boiler.
                                     135

-------
                          30
       35        40
DRY SOLIDS FIRED, 1()3 Ib/hr
                                                       45
                                                              2.0
         Figure 3-48.   Effect of solids firing rate on reduced sulfur
                  emissions and steam generation efficiency.l              '

      Based on computer models of recovery boiler operations, the particulate
 emissions  (gains  per  dry standard cubic foot) increase sharply with increases
 in  the  heat value and solids content of the black liquor (Figures 3-49 and
 3-50).   These changes are primarily the result: of changes in the heat produc-
 tion  rate  of the  char bed.54
      Because the  heat value and solids content of black liquor are dependent
 on  a  number of process  variables in the pulping process (e.g., digesters,
 evaporators,  wood species,  and  harvest conditions),  day-to-day firing condi-
 tions and  liquor  properties  may vary significantly.
      Increased  firing rates  in  the  recovery furnace  can increase flue gas
 flow rates.   The  amount  of flue gas produced by the  combustion of a  specific
 black liquor  can  be calculated  by developing a  factor  similar  to an  F-factor,
which is defined  by the  elemental composition of the liquor.   The F-factor is
 the measured  flue gas volume  (dry standard  cubic feet  per minute corrected to
 0% flue gas oxygen) divided by  the  boiler black  liquor solids  firing  rate
 (pounds Bis/minute).  The value  of  the  F-factor  varies  from mill  to mill

                                      136

-------
   600


.-, 500
M
-H
i
£ 400

I
I 300
01

a* 200


   100
                         J_
                                                          S.O
                                                          4.5
                                                          4.0
                                                          3.S
                         58         62         66        70
                       Black Liquor Solids Conwntration, ft
                                                          3.0
Figure 3-49.   Effect of  black  liquor solids concentration.
                                                           54
            2000
                                                        - 2
              5200         5600        6000       6400        6800
                    HMting Value of Solids to Rirnac*, Btu/lb

    Figure  3-50.   Effect of black liquor heating value.
                                                       54
                                  137

-------
because of the variation in species, pulp yield, makeup chemicals, and evaporator
                                                                            CO
operation.  This value, however, is reasonably constant for a specific mill.
     Knowledge of the F-factor allows periodic calculation of flue gas
volumes to ensure that the boiler is not exceeding design values.   For a more
accurate calculation of velocities in the combustion chamber and the total
flue gas volumes, the F-factor must be corrected for water evaporated from
the fired liquor, water of combustion, temperature, excess air, and miscel-
laneous additions such as steam from soot blowing.
     Estimates of uncontrolled emission rates as a function of furnace firing
rate are boiler-specific.  Generally, the uncontrolled emission rate of a
typical indirect-contact, recovery boiler is 8.0 gr/dscf.  A baseline uncon-
trolled emission rate for use in making future comparisons generally can be
obtained from the vendor's acceptance performance stack test on the ESP.  In
summary, increased firing rates increase both the flue gas volume and the un-
controlled particulate emission levels.
     Char Bed Temperature—As the firing rates of a recovery boiler are in-
creased, the temperature and diffusion conditions in the char bed tend to in-
crease.  This in turn leads to an increase in uncontrolled particulate emis-
sions from the boiler.  A qualitative discussion of the mechanisms by which
particulate emissions are increased is presented in the remainder of this sub- ;
section.  Dehydrated liquor (char) on the furnace hearth is burned at a high
temperature to allow the inert portion of the liquor to be melted and drained
from the hearth.  A mixture of sodium sulfide and sodium carbonate must be
maintained under reducing conditions to prevent oxidation to sodium oxides.
     Under normal operation, elemental sodium is vaporized and reacts to form
Na?0.  The rate of evaporation depends on the char bed temperature and the
diffusion conditions in the smelt zone.  As the sodium evaporates from the
bed, it reacts with oxygen in the primary air zone to form Na20.  The Na20
reacts with C02 to form Na2C03.
                               55
     The char bed temperature also determines the rate of sulfur released to
the flue gas.  Sulfur is commonly present in the flue gases as S, H2S, S02, or
S03.  The higher temperatures favor the formation of S02 and S03-  Sulfur in
the form of S and H2S reacts with excess oxygen in the oxidizing zones of the
furnace to form S02 and SOg.  The Na2C03 reacts with the S02 to form Na2S03,

                                      138

-------
which  is  later oxidized to Na2S04.  Typically,  the  Na2$04 deposits on the heat
exchanger surfaces (screen tubes, superheater,  and  boiler tubes) and must be
removed through continuous sootblowing.  These  deposits  reduce the heat transfer
and  the overall efficiency of the boiler.  This  increases the gas temperature
entering  the ESP,  as well  as the superficial velocity.   As a  result, there is
a tendency to overfire the boiler to achieve the required steam flow.
     Computer models of smelt bed temperatures  indicate  that  the smelt bed
temperature is a function  of total combustion air and primary air.   Figures
3-51 and  3-52 show smelt bed temperature as a function of total  and primary
                   54
air, respectively.
                  **
                  I
                      so
                      45
                      40
                      35
                      30
                                                          2000
                                                          1900
 1800
                       100     110    120     130     440
                            Total Air to Uiit, parent of theoretical
                                                          1700
150
          Figure 3-51.  Bed temperature  as  a  function of total air.
                                                                    54
                             Priauy Air, fnotat of total
                                                           1600
                                                          60
        Figure  3-52.   Bed temperature as a function of primary  air.
                                                                     54
     Boiler Excess Air—The amount of combustion air supplied  to  the  furnace
influences the chemical-thermodynamic equilibrium at the smelt bed.   This
determines the amount of  sodium that is volatilized and ultimately  lost from
                                      139

-------
the smelt bed.  Figure 3-53 shows the theoretical loss of particulate from
the furnace as the total air to the furnace is increased (primary air is fixed
at 45%).54
          12
        S 10
        1  S
        I

        I
                   1500
                  1000
                8
                I
                a"
                   500
                        \H»2304 in Fun
100

 80 *fc
 60 *^.
   a
 40
          Figure 3-53.  'Effect of total air supplied to the unit.
 6  ;TJ

 :j
  tst
 3  "*
 2  f
 1  |
 0  *

54
     The amount of boiler excess air needed for complete combustion is
normally between 110 and 125 percent of theoretical air (stoichiometric air).
A minimum level of excess oxygen (1 to 2%) must be maintained to ensure com-
plete combustion and reduce the formation of
                                                 12
                                                     When the amount of excess
air is above 125 percent (5% 02 in the flue gas), the formation of S03 in-
creases.59  The formation of S03 and H2S04 can be increased considerably if a
residual fuel oil with a high content of vanadium pentoxide (V205) is fired in
combination with the black liquor.  The SO, is absorbed in the particulate
                                                        12
at low temperatures, which makes the particulate sticky.    This sticky par-
ticulate fouls heating surfaces in the economizer and reduces heat transfer
rates.  The deposits may result in a high draft across the economizer.  The
particulate also causes severe operating problems when collected on the plates
of the ESP.  The sticky salt cake (particulate) cannot be removed effectively
by boiler soot blowers or removed from the ESP plates by normal rapping in-
tensity.60
     Higher than normal excess air increases the volume of flue gas, which re-
sults in increased furnace vertical velocities and reduced heat transfer ef-
ficiencies.  If the high excess air is caused by an increase in primary air
in the char bed zone, sodium evaporation will increase along with the particu-
late emission rate.  When the high excess air is caused by increased secondary
air, the formation of S03 in the flue gas is more likely.
                                      140

-------
     In summary, the firing of  the boiler at  high excess air  has  three  effects:
1) increased particulate emission rates on an  uncontrolled  basis, 2)  in-
creased flue gas volume to the  ESP, and 3) formation of S03 that  reduces  ESP
power as a result of salt cake  buildup on both the plates and the wires.
     Primary Air--Primary air is required to provide complete combustion  and
to maintain the temperature in  the char bed to prevent a condition called
"blackout."  The amount of air  is a compromise between maintaining sufficient
combustion and reducing abnormally high vertical velocities in the furnace.
The total air volume (secondary plus primary)  must be high  enough to  produce
complete combustion, but be limited so as to reduce the vertical velocity in
                                      12
the furnace and total flue gas volume.    A higher velocity increases the
release of sodium and sulfur from the char bed because of the increase
in diffusion of the vapor from the bed.  The higher volume  also increases the
combustion rate of char, which increases bed temperature.   Figure 3-54  shows
the theoretical loss of particulates as a function of percentage increases in
primary air.  This produces an accumulation of deposits on  the heating  surface
of the boiler (after cooling of the gas stream and condensation of fume).  This
accumulation causes an increase in the particulate emissions.
           "
           12
           10
            *
        i  «
        I  2
   •  1500
   i

   3  1000
.6
  I  500
  8
  3
100
 80
 60
 40
 20
 0
  Figure 3-54.  Theoretical loss of particulate as a function of percentage
                         increases in primary air.
                                                  5
-------
     Secondary Air—The total amount of primary and  secondary  air  required
for combustion is between 110 and  125 percent of  theoretical air  (stoichiometric
air).  The normal limits are between 2 and 5 percent excess 02-
     The secondary air should be a minimum of 40  percent of the total  air
(maximum of 65 percent of the theoretical air).12 In  boilers  with a  high char
bed, the secondary air has two purposes.  The primary  purpose  is  to complete
combustion of CO gas released from the char bed as it  moves up the furnace
walls.61  The secondary purpose is to provide primary  air  in the  center  of
the furnace to burn the char bed.
     Primary Air Temperature—The  primary air temperature  has  a direct in-
fluence on the smelt bed temperature.  An increase in  the  primary air temperature
has the same effect as an increase in the primary air  volume.  Figure 3-55
shows the effect of primary air temperature.54   In modern  designs, the primary
air is heated with indirect steam  to approximately 400°F to  increase  the bed
temperature.  This reduces the TRS emissions from the  bed  and  increases  the
emission of Na2S04.50  Changes in  primary air- port design  are  also used  to
increase velocities, which help to stabilize bed  temperature through  better
air penetration.
                   600

                   soo

               3
               J  <°°
               I
               ft  300
               I
               a  **>
               rf*
                   100
                                        I
                                     1   r
100       200      300       400
          Priauy Air T«pwatux«, *F
                                                        500
              Figure 3-55.   Effect of primary air temperature.
                                                              54
                                      142

-------
      Sulfur-to-Sodlum  Ratio—As the ratio of  sulfur to sodium in black liquor
 increases, the amount  of sodium in the smelt  that  is available to combine
 with sulfur gases  is reduced.   This results in an  increase in S02 and TRS
-emissions and a decrease in sulfidity across  the furnace.   As a result, the
 smelt has a lower  sulfidity (Figure 3-56).  At sulfidity levels in excess of
                                                                                en cfl
 30 percent it is difficult to  comply with applicable TRS emission standards.   '
             1
                8000
                6000
              I
              jf 4000
              M
              n
              I
              sf 2o°°
                                                               8  3
6 S.
  §
  r
4 y
                100
             if
             53
             I
                 80  -
                 60
                 40
                 20
                                        — — Reported,by Fukui
                         0.2    0.4    0.6    0.8    1.0    1.2
                              Molar Ratio S/Jtaj in Black Liquor
                                                             1.4
      Figure 3-56.  Effect of  sulfur-sodium ratio in  the  black liquor.
                                                                        54
                                        143

-------
     Chlorine  in  Black  Liquor—Chlorine can enter the pulp process through
wood-born contaminants,  chemicals,  or mill  wastes.   The most common sources
are saltwater  log storage  and  bleach plant  effluent.   The presence of small
amounts of chlorine  increases  the loss of particulates from the smelt bed.
The chlorine combined with sodium (NaCl) is highly volatile.  Figure 3-57 shows
the theoretical loss of  sodium chlorides as a function of percent chlorine in
                  54
the fired liquor.    The loss  of HC1  from the furnace can increase corrosion
in the ESP and ductwork  if significant cooling occurs as a result of inleakage
or improper insulation.
                ~ 800
                .3
                a
                  600
                I
                § 400
                •H
                K
                  200
                s
                   40
                   30  -
                   20  -
                   10  -
                     01234
                         Oilorin* in Black Liquor, percent (w»t buis)
             Figure 3-57.  Effect  of  chlorine in black liquor.
                                                               54
                                      144

-------
3.3.1.3  Control —
     The primary method of controlling participate emissions from recovery
furnaces is electrostatic precipitation.   At one time venturi scrubbers were
used for primary and secondary particulate control on some designs;  however,
since their use has been discontinued, the following discussion is limited to
ESP's.
     Equipment Design—The three basic processes in electrostatic precipita-
tion are 1) the transfer of an electric charge to suspended particles in the
flue gas, 2) the establishment of an electric field for removing the particles
to a collecting electrode, and 3) removal of the particles from the  ESP with
as little loss to the atmosphere as possible.  Figure 3-58 illustrates the
                                                        fi?
basic processes involved in electrostatic precipitation.
     Particulate matter is collected in an ESP by means of an electrical
charge placed on the particles and a grounding surface of opposite charge.
The particles move in an electrostatic field created by electrodes operating
at a negative potential ranging from 30,000 to 80,000 volts.  The electrodes
consist of wires placed perpendicular to the gas flow between collecting plates.
As the potential on the electrodes is increased, electrons are released that
charge the particles passing between the wires and plates.
     Depending on the size of the wire and its roughness, shape, and dust
coating, changes occur in the voltage at which the electrons begin to flow
(as measured by secondary current).  The cloud of electrons surrounding the
electrode is called a corona, and the voltage at which current occurs is de-
fined as the corona initiation voltage.
     The charged particles move to the collecting plates and bleed their charge
through the dust layer.  The rate of release of this charge is indicated by
the secondary current.  When the dust concentration in the gas stream is low,
the relationship between secondary voltage and secondary current is an expres-
sion of the electrical resistance of the system.
     The secondary voltage and resulting secondary current may be increased
until the potential is high enough to allow a spark to occur directly through
the gas stream to the plate.  This results in power loss without the charging
of particles.  In older ESP designs, the optimum utilization of power (highest
efficiency) occurred at moderate spark rates (>50/min).  With the use of new
                                      145

-------
      FREE
   ELECTRONS
 REGION OF\
CORONA GLOW\
                                   e
                                 jELECTRONS
     CORONA  GENERATION
                                           ELECTRON
                                     DUST
                                   PARTICLE
                                      GAS
                                    MOLECULE
                              CHARGING
                            -© WIRES
                           TURBULENT
                           GAS  FLOW
                                                           >>\  RAPPING
                                                            N   SYSTEM
         COLLECTION
                                              COLLECTING
                                              PLATE
                                             ;HOPPER
                                                ASH REMOVAL
                                                SYSTEM
                          REMOVAL
Figure 3-58.   Basic  processes involved in electrostatic precipitation."
                                                                     62--

-------
digital control systems that sense the occurrence of sparking, optimum power
levels can be maintained without sparking or at a moderate spark rate (<50/min).
                                                         fi?
     Figure 3-59 shows the general arrangement of an ESP.    The space between
the plates is called the gas passage or lane.  Although the spacing between
plates varies with manufacturer, it is typically between 9 and 10 in.  The
electrodes (wires) are placed at equal distances between the plates (at 8-
in. intervals in the direction of gas flow).  The wires may be individually
supported and tensioned by use of weights or rigidly held in a frame.  The
electrode cross section may be round, square, twisted, 'barbed wire, or strained
barbed wire.   The shape and diameter determine the corona initiation voltage.
     In general, the resistance of the particle to conduct its charge to the
collecting surface is defined as the particle resistivity.  The high moisture
content of the gas stream from recovery furnaces results in a highly con-
ductive particle (low resistivity).  This low resistivity allows the particle
to release the charge quickly at the plate and to be effectively removed from
the plate with moderate plate rapping.
     Power is supplied to the ESP by rectified high-voltage transformers.
The rectifiers may be full-wave or double haIf-wave design, depending on the
                                                     63
sectionalization (chambers) of the ESP (Figure 3-60).
     The ESP is divided into separate collection surfaces arranged in series
to-allow increases in power levels as dust is removed from the gas stream and
to allow individual field rapping.  The total power input to each successive
field increases until the transformer-rectifer (T-R) set limit is reached.
Power usage by each section (field) is indicated by the product of primary
voltage and current of the transformer and by the product of secondary
voltage and current of the rectified power supply.  The primary power level
is not considered a true measurement of ESP power because of inefficiencies
and losses in the transformer and rectifier.  Although the secondary power
readings indicate the power delivered to the ESP, in some cases they do not
measure the power delivered to the particle charging mechanism because of
insulator tracking, distribution  losses, etc.
     Because of the nature of the charging process, the collection efficiency
is affected by the length of time the particles are in the electrostatic
field  (time of treatment).  Because the physical dimensions of the ESP are
fixed, the time of treatment is determined by the total gas flow rate, box
                                      147

-------
                                      GROUND  SWITCH BOX-
                                        • ON TRANSFORMER
         TRANSFORMER-
         RECTIFIER
HEAT JACKET
PERFORATED
DISTRIBUTION
PLATES
 DISCHARGE
 ELECTRODE
DISCHARGE
ELECTRODE
VIBRATOR
  COLLECTING
  ELECTRODE
  RAPPERS
                                                                      TOP HOUSING
                                                                      ACCESS DOOR
                                                                      TOP HOUSING
                                                                      HOT ROOF
        CCESS DOOR
       BETWEEN
       COLLECTING
       PLATE  SECTIONS
                                                     COLLECTING ELECTRODES
                                         WET BOTTOM
       Figure 3-59.   Typical wet-bottom ESP with  heat jacket.62

                                        148

-------
    R.-L
Low
Voltage
Input
                   High Voltage
                   Transformer
                  —s*.
Bridge
Rtetifitr (Static)
One or More
But Section*
                        FULL-WAVE CIRCUIT SCHEMATIC
 Usually
 Linear Reactor
                                                        Two Separate Bus Sections
                                                        or Fields
Low
Voltage
Input
                   DOUBLE HALF-WAVE CIRCUIT SCHEMATIC
      Figure 3-60.    Electrical  diagram  for ESP  T-R set.
                                                                       63
                                      149

-------
                                       65
 length (number of fields),  and cross-sectional  area;  i.e.,  the  time  of  treat-
 ment is defined by the superficial  velocity between the  plates  and the  box
 length.   The  deviation from the average  velocity  is limited to  +10 percent  of
 the mean value.
      The design amount of power to  be  delivered to the ESP  in each field  is
 sometimes specified in terms of the length  of  the wire electrodes.   The inlet
 field is typically sized  for 0.2 milliamp  (mA)/ft of  wire,  and  the outlet
 field may be  sized for up to 0.10 mA/ft  of  wire.    Accordingly,  the inlet  field
 T-R sets are  sized to  be  consistent with anticipated  power  consumption.   The
 plate area  used per 1000  acfm of gas being  treated is one convenient measure
 of  the potential  collection  capability of the  ESP.  This term is  defined  as
 the specific  collection area in square feet/1000  acfm.
      There  are two methods  of supporting the discharge electrode  in  recovery
 boiler ESP's.   In  the  first  method,  referred to as the weighted-wire design,
 each electrode is  individually supported and tensioned between  the plates
 (Figure  3-61).     In the  second method,  referred  to as a rigid-frame design,
 the electrode is  attached to a rigid frame  between the plates (Figure 3-62).
 A modification of the  second design  is a rigid-pipe electrode system in which
 the corona  is  generated on  the tip  of  spikes attached to a  vertical  pipe.
      Collected particulate can be removed from  the ESP in three ways.   In the
 first method,  referred to as  a wet-bottom ESP,  the salt cake is allowed to
 fall  into an  agitated  pool of black  liquor  in  the bottom of the ESP  (Figure
       65
 3-63).    In the second method,  referred  to  as  a dry- or drag-bottom  ESP,  the
 salt cake is  allowed to fall  onto the  flat  bottom of  the ESP shell,  where a
 drag chain  physically moves  the material to a  discharge screw (Figure 3-64).
 The  third method of dust  removal consists of a  pyramid-shaped hopper with
 rotary air  locks and slide-gate discharges.  This design is not often used  in
 recovery boiler ESP designs  because  the  hopper  tends to plug.
     To  remove the  particulate from  the  discharge electrodes and  collection
 plates,  the ESP collecting surfaces  are  vibrated  or rapped.  Rapping equip-
ment falls  into three main categories:   electrical, physical, and pneumatic.
The  electrical rappers consist of MIGI (magnetic  impulse gravity  impact)
 rappers  and vibrators.  The  physical rappers are  rotating falling hammers,
 either internal or  external  to  the  ESP.  The pneumatic rappers are air
                                       65
150

-------
Figure 3-61.  Typical weighted-wire ESP with drag bottom.
                                                         65
                            151

-------
Figure 3-62.  Rigid-frame design.65



               152

-------
Figure 3-63.  Wet-bottom ESP.
             153
                             65

-------
IQ
 ro
 00
 en
 o
 -s
Id
 O
 Q>
 _l.
 3
 fD
 cr

-------
cylinder hammers.   Figure 3-65 shows a MIGI rapper, and Figure 3-66 shows a
typical internal falling hammer rapper design.
     Buildup on the plates resultsS1n low collection efficiency due to re-
duced power input to the ESP.  The effect is most prevalent when the flue gas
temperatures are below 300°F and a combination of high boiler excess air and
ambient air inleakage is occurring.
     In general, ESP control efficiency is determined by the initial design
of the unit and the operating characteristics of the source.  The major design
factors include plate area, superficial velocity, and sectionalization of the
unit and size of the T-R set.  Figure 3-67 shows the relationship between
                                                             O
design efficiency and the design specific collection area (ft /1000 acfm) for
modern ESP's.66  It should be noted that increasing the plate area above 400
ft2/1000 has a diminishing effect with regard to improving performance, pri-
marily because of the predominance of factors not related to corona power,
such as hopper sneakage and rapping losses.  To minimize the effects of rap-
ping loses and to improve residence time, manufacturers have been designing
the newer ESP's with lower superficial velocities.  The accepted maximum
velocity in current technology is about 3.5 ft/s.    Figure 3-68 shows the
design velocity of 20 randomly selected units.
     It should be noted with  increased treated gas volume, an  increase in gas
temperature can increase the  superficial velocity beyond design values and
reduce the effective SCA and  collection efficiency.
     Instrumentation—Optimum performance of an  ESP depends on effective  con-
trol of the operating parameters  that  can vary because of changes  in  the
physical characteristics of  the ESP or the  flue  gas stream  being treated.
Recent ESP  installations generally are equipped  with instrumentation  for  moni-
toring and  recording the major operating parameters.  The operator and  in-
spector should  thoroughly  understand  the function  of each instrument  and
associated  records  to evaluate ESP  performance.
     Instrumentation for a  kraft  pulp  recovery boiler  ESP generally consists
of monitors for ESP power  input,  flue gas  temperature, oxygen  content,  and
opacity, rapper operation,  and the particulate discharge  system.   The in-
struments  are  generally located  in proximity to  the ESP  unit.   When a plant
has  more than  one  ESP,  the instrumentation for all  the  ESP  units  can  be housed
in a centrally located  control.
                                       155

-------
Figure 3-65.  MIGI rapper cross section.
65
                    156

-------
Figure 3-66.  Internal falling-hammer rapper design.
                         157

-------
   100
    99
o
    98
    97
   96
                      DEUTSCH-
                      ANDERSON
                                 MATTS-OHNFELDT
                     /

                 100
0
200         300          400        500	600

 DESIGN SCA, ft2/1000 acfm
       Figure 3-67.   Design SCA and efficiency of 20 recovery boiler ESP's.66
                                        158

-------
                                               6SI
                                   SUPERFICIAL  VELOCITY, ft/s
    vo

    tn
                                                  oo
                                                 —r
10
 -5
 fD
 CO
 CT>
 CO
 GO
 £=
•O
 n>


 I?.
 o

 O)'
fD

o
o
*1,

4?


fD
       5
       PO
CO
    vo
    CJ1
    O
           en
           C7I
    en
    o
    tn
£=
CO
fD
O)
-S
fit
    vo

    o
           vo

           tn
                                                   •   •
fD
Q.
 CT>
 CT>
           vo
           oo
           o

-------
      In the ESP instrumentation block diagram shown in Figure 3-69,62 the
 ESP has four T-R sets, each of which has two bus sections.  Primary voltage
 and primary current are measured for each T-R set.  Secondary voltage, secon-
 dary current, and spark rate are measured for each bus section.   Figure 3-70
 shows the positions of various instruments in an ESP circuit.   The ESP in this
 figure has four primary voltmeters and four primary ammeters;  the secondary-
 side instrumentation consists of eight secondary voltmeters,  eight secondary
 ammeters,  and eight spark rate meters.
      The function of each instrument determines its relative  location in  the
 circuit.   The primary ammeter, for example,  is always  located  ahead of the
 transformer to indicate the current available for transformation.   Temperature
 is measured at the ESP inlet,  and opacity is measured  at  the outlet.
      Primary-side meter readings only indicate the general  condition  of the
 ESP and cannot be used for complete diagnosis of internal  conditions  while
 the ESP is in  operation.   Nevertheless,  an experienced operator  generally can
 use the primary readings  to indicate wire breakage,  severe  plate buildup,  and
 discharge  problems.
      Energization  power for an ESP  is supplied  at the  primary  side  of the T-R
 set.  The  alternating  current  electrical  power  is normally  220 or 460 volts.
 The voltmeter  dial for primary voltage shows  a  range of 0 to 480 volts.
 Normal  operation  is  usually between  220  to 460  volts.   A temporary  deviation
 of +5 percent  from the rated supply  voltage  is  fairly  common.  The  primary-
 side voltmeter  is located  ahead  of the T-R set,  but after the  power control
 circuit, linear reactor, and feedback network.   This positioning ensures
 measurement of the regulated voltage  available at the  T-R set.
     The ammeter on  the primary  side  indicates the current drawn by the T-R
 set.  The  primary ammeter  is located  between  the  T-R set and the power con-
 trol circuit, linear reactor, and feedback network.
     Secondary-side  instrumentation indicates the power input  to an individual
 ESP section.  The secondary-side instrumentation  usually consists of a volt-
meter, an ammeter, and a spark rate meter for each bus section.  Secondary-side
meters are used to produce current/voltage relationships that characterize the
internal conditions of the ESP.  This allows maintenance personnel to concen-
trate on the most severe problem areas.
                                     160

-------
CTl
                                                                                         g3E=3E=)E^
                 SI     SR     PV     PI    SV    SI     SR
SV     SI     SR     PV      PI     SV     SI     SR
                                                                                                       ACITY
                                                                                                     INDICATOR
                       IINLET CAS TEMPERATURE INDICATOR
                                 SV:  SECONDARY VOLTAGE
                                 SI:  SECONDARY CURRENT
                                 SR:  SPARK RATE
                                 PV:  PRIMARY VOLTAGE
                                 P!:  PRIMARY CURRENT
                                     Figure 3-69.   ESP instrumentation  diagram.
                                                                                   62

-------
         4M V
         i w
         CO HI
-o  o
                    CONTROL
                     TRANS.
cr>
ro
                                                AUTOMATIC CONTROL HOOULC
                                                               H.V.
                                                   I VOLTAGE )  TRASSF.
                                                   \NETER
ESP
                                       Figure  3-70.   Positions  of measuring  instruments.
                                                                                                68

-------
     Secondary voltmeters are calibrated in kilovolts to measure the high
voltage of the power input to the discharge electrode.  The secondary volt-
meter is located between the rectifier output side and discharge electrodes
to indicate the direct current voltage across the discharge electrodes.
     The secondary-side ammeter indicates the current that is being supplied
to the discharge electrodes in one section of the ESP.  The secondary current,
which is produced by stepping down the primary current in the transformer,
is measured in mi Hi amps.
     The spark rate meter is a major indicator of ESP performance.  The spark
rate indicates the rate of sparks in a single ESP section.
     Certain instruments external to the ESP are also important in the over-
all operation of the ESP.  These include opacity monitors, oxygen monitors,
inlet gas temperature indicators, and discharge system monitors.
     Opacity is limited under the NSPS regulations for kraft pulp mills, and
new kraft pulp mills are required to install and operate continuous opacity
monitors.  A correlation can be developed between opacity and mass emissions
during the performance test that allows the operator to use opacity as an
indicator of overall performance with respect the mass emission rate.
     The level of oxygen in the flue gas is an indication of excess gas
volume due to boiler excess air and air infiltration into the ductwork and
the ESP.  Because of the relationship between ESP performance and total gas
volume, high oxygen content in the flue gas is an indication of reduced
performance.   The optimum levels-of excess oxygen should be identified during
the performance test and used as a baseline for determining when maintenance
may be needed.
     Temperature is usually measured at the exit of the recovery boiler.
The temperature of the flue gas affects corona power and ESP efficiency; i.e.,
low temperatures can cause acid and moisture condensation, which lead to
corrosion and eventual  structural  failure.
     Instrumentation associated with the discharge system varies with the
type of discharge system being used, i.e.,  wet-bottom or drag-chain.   The
instrumentation for wet-bottom ESP's usually consists of a float to indicate
the level of the black liquor in the wet bottom and a flow meter to indicate
volumetric flow.   Some mills also have .ammeters to indicate whether the pumps
                                     163

-------
that  transport the  black  liquor are working properly and tachometers to
indicate the revolutions  per minute of the agitators.
      For drag chains,  the only two parameters that are usually measured are   •
the revolutions per minute of the drag chain and the current to the motor
that  operates the drag chain.  These  instruments are basically used to tell
whether the drag chain is operating properly.  Too much tension on the drag
chain can cause the chain to break, and too little tension can cause particu-
late  to build up in the bottom of the ESP.  If the particulate continues to
build up, it can cause the chain to break.  In addition, significant buildup
that  goes unnoticed for a period of time can eventually ground out the system.
      The basic instrumentation associated with rapping is a meter to indicate
the rapper has "fired."   Information on the internal operation of the rapper
and rapper intensity is extremely difficult to obtain without shutting down
the ESP.
3.3.1.4  Boiler Malfunctions-
      Malfunctions that increase emissions can be divided into two areas:
1) those that occur as a  result of furnace operation, and 2) those that occur
as a  part of control equipment operation.  Several malfunctions that result
from  improper boiler operating practices also may have an impact on ESP
performance.
      The following  is  a summary of the major boiler operating or design
parameters and associated malfunctions that may Increase emissions.
      Firing Rate--The  firing rate of the boiler in pounds of black liquor
solids per hour defines both the heat input to the boiler and standard cubic  •
feet  of flue gas volume generated.  The flue gas volume and heat input change
with  the chemistry  and the heat content of the liquor.  At defined chemistry
and heat values, predictable volumes of a flue gas may be calculated.  The
F-factor for black  liquor varies from mill to mill because of species, pulp
yield, and makeup chemicals, but is reasonably constant for specific mills.
An increase in firing  rates above design (flue gas volume) results in increased
vertical gas velocity  through the combustion zone of the furnace and an in-
crease in particulate  emissions due to the entrainment of black liquor droplet's
12
and char particles and increased char bed temperature.
                                                      1,53
                                     164

-------
     Participate emissions also increase at high firing rates because of the
increased volume of primary air.  The release of sodium in the flue gas from
the char bed increases as the flue gas volume increases as a result of favor-
                          54
able diffusion conditions.    The release of sodium is related to the tempera-
                                                           CO
ture of the bed and is primarily the result of evaporation.    The rate of
evaporation depends on the diffusion conditions (gas velocity) in the zone
between the flue gas and the char bed.
     The firing rate has a dual effect.  It influences the flue gas volume
treated by the ESP and the uncontrolled particulate emission rates.
     Char Bed Temperature — The char bed temperature appears to have an effect
on the evaporation rate of sodium to the flue gas.  The amount of sodium that
is distributed to the flue gas increases sharply with temperature, following
                                          12
a curve similar to a vapor pressure curve.
     The sodium evaporates from the bed as elemental sodium reacts with oxygen
in the primary air zone to form Na^O.   The Na^O reacts with C02 to form Na-pCO
The amount of sodium released from the bed does not appear to be dependent on
the sodium content of black liquor solids.
     The temperature in the bed also determines the release of sulfur to the
flue gas.  The sulfur is commonly present in flue gases as S, H2S, or S02.
The lower temperatures favor the formation of S and H2S.  Typical char bed
temperatures are 1700° to 2200°F.54
     The sulfur that is released reacts with excess oxygen in the oxidizing
zones in the furnace to form S0.  The
reacts with the S02 to form
                               2.        ^
Na2S03 that is later oxidized to Na2S04.  '
     It should be noted that in some designs the combustion of the char may
be accomplished on the boiler, wall.  In these designs no char bed (profile)
is maintained, and the smelt drains from the walls onto a sloped hearth for
removal.
     Primary Air — The primary air is required to provide complete combustion
and to maintain the temperature in the char bed to prevent a condition called
"blackout."
     The amount of air is a compromise between maintaining sufficient combus-
tion and reducing abnormally high vertical velocities in the furnace.  The
increased velocity results in an accumulation of deposits on the heating
                                     165

-------
 surface  of the  boiler  (after  cooling  of the  gas  stream and condensation of
 fume).   This  accumulation  causes  an increase in  the  particulate emissions.
     The increased  rate  of primary air  (particularly at  high velocities)  in-
 creases  the release of sodium and sulfur from the char bed because of a
                                                          rq
 favorable increase  in  diffusion of the  vapor from the bed.    The increased
 volume also increases  the  combustion  rate of char, which increases bed
 temperature.  Particulate  emissions increase sharply when the amount of
 primary  air exceeds 45 percent of the total  air  volume.
     Prolonged  operation at low primary air  volumes  can  result in increased
 char bed height, which must be reduced  by an increase in bed temperature.
 The most common method of  reducing the  bed height is to  increase the primary
 and secondary air volumes.  This  can cause the smelt ratio to be altered  as
 oxidation and temperature  conditions are changed.
     The increased  combustion  can increase the flue  gas  volume from the
 boiler to a point where  it is  greater than the amount calculated from the
 instantaneous firing rate  of  the  boiler.  For the volumes to be equal to  in-
 stantaneous firing  conditions, the char bed  combustion rate must be at equi-
 librium  with  the amount of char deposited on the bed.  When the bed is being
 reduced  in  height,  the flue gas and particulate  emission rates will be greater
 than those  predicted by the black liquor firing  rate.
     Secondary Air—The total  amount of primary and  secondary air required
 for combustion  is 110  percent  of  theoretical  air (stoichiometric air).  The
 normal limits are between  2 and 5 percent excess Oy-
     The  secondary  air should  be  a minimum of 40 percent of the total air
                                               1 9
 (maximum  of 65 percent of  the  theoretical air).    In boilers with a high char
 bed, the  secondary  air has  two purposes.- The primary purpose is to complete
 combustion of CO gas released  from the  char  bed as it moves up the furnace
walls.  The secondary  purpose  is  to provide  primary air  in the center of  the
furnace  to burn the char bed.
     The  total air  volume  (secondary plus primary) must  be high enough to
produce complete combustion, but  it must also be limited to reduce the verti-
cal velocity  in the furnace and total  flue gas volume.
     Boiler Excess Aii—The amount of boiler  excess air  needed for complete
combustion is typically between 110 and  125  percent of theoretical  air
                                     166

-------
(stoichiometric air).  When the amount of excess air is above 125 percent (5%
Og in flue gas), formation of S03 increases.  '    The S03 is absorbed in the
particulate at low temperatures, which makes it sticky.  This sticky particu-
late fouls heating surfaces in the economizer and reduces heat transfer
rates.  The deposits may result in a high draft across the economizer.  The
particulate also causes severe operating problems when collected on the
plates of the ESP. ,   The sticky salt cake cannot be removed effectively by
boiler soot blowers or removed from the ESP plates by normal rapping inten-
sity.
     Buildup on the plates causes low collection efficiency as a result of
                                                                              i
reduced power input to the unit.  The effect is most prevalent when flue
gas temperatures are below 300°F and a combination of high boiler excess air
and ambient air inleakage is occurring.
     This condition can be identified in the ESP when high secondary voltage
(> 50 kV) and low secondary current (< 100 mA) are observed in the inlet fields.
Figure 3-71 shows a typical secondary current pattern for a unit experiencing
salt cake buildup as a result of high excess air (SO, formation).
     The formation of SO, and H2S04 can increase considerably when a residual
fuel oil with a high content of vanadium pentoxide (V90r) is fired in combi-
                             70
nation with the black liquor.
     The firing of the boiler at high excess air has three effects:  1) in-
creased particulate emission rates; 2) increased flue gas volume to the ESP;
and 3) formation of SO., which reduces ESP power because of salt cake buildup
on both the plates and the wires.
3.3.1.5  ESP Malfunctions--
     The following is a discussion of the major operating and design param-
eters and associated malfunctions that have an impact on ESP performance.
     Operation and maintenance (O&M) of ESP's is a broad subject involving
all aspects of ESP performance.  It cpvers all component parts and all oper-
ating conditions.  In general, maintenance is considered to be the routine
analysis and replacement of components parts that have failed because of age
or abuse.  Maintenance requirements may be increased as a result of poor
operating practices or reduced through superior system design.  Detailed and
exhaustive maintenance practices do not, however, necessarily yield superior
or exceptional ESP performance.

                                     167

-------
    600
    500
    400
LU
o:
O


or
«x
o

o
CJ
LU
C/7
    300
200
   100
             SOUTH
                                2

                              FIELDS
    Figure 3-71.   Typical secondary current pattern for unit
                experiencing salt cake buildup.71
                               168

-------
      Poor design  or  operating  practices  that  cause  a  high  level of maintenance
 activity are  not  necessarily corrected by  that maintenance activity.   Because
 the  symptom is  often treated,  rather  than  the real  problem,  this discussion
 centers  around  good  operating  practices  and major design problems instead of
 routine  maintenance  activities.   It is imperative,  however,  that a good
 inspection and  diagnostic  program be  used  to  identify problem areas and to
 direct corrective action.  Such action may include  changes  in operating
 practices, the  basic system of design, and maintenance schedules.
      In  general,  the ESP recovery boiler system must  be operated within the
 scope of its  design  variables.  Deviation  from the  initial  design for  inlet
 temperature,  inlet grain loadings, gas volume, superficial  velocity, corona
 power, and velocity  distribution  can  adversely affect expected performance
 levels.   The  following discussion deals  with  the importance  of several oper-
 ating parameters  and the major causes of poor performance.
      Gas  Volume—The gas volume treated  by an ESP is  defined by the boiler
 size  (heat input), excess  air, temperature of the gas stream, inleakage
 through  duct  flanges  and across cascades,  and black liquor composition.
 Several  of these  variables, e.g.,  black  liquor composition, are fixed over a
 very  narrow range.   The other  parameters have a wider range of variability
 and depend on boiler  operation, boiler age, or lack of maintenance.
      Changes  in those variables that  increase the gas volume treated by the
 ESP decrease  ESP  removal efficiency.  Because the cross-sectional  area of the
 ESP is fixed, any increase in  gas  volume causes a corresponding increase in
 superficial velocity and a decrease in SCA.  Performance of the ESP has been
 improved  in recent years by increases in SCA through  the use of more fields.
The SCA has also been increased by increasing the cross-sectional  area of the
box to reduce superficial velocity.  Currently,  ESP1s are being designed with
SCA's in excess of 400 ft2/1000 acfm and superficial velocities near 3.5
ft/s.  Increasing the operating flue gas volume above the design effectively
reduces the size of the ESP and reduces  its performance.
     The amount of flue gas generated by a pound of black liquor solids is a
function of the elemental composition of the solids.  The organic components
are carbon, hydrogen, oxygen,  and sulfur.  The nonorganic or ash portion of
the solids is also important because it combines with the oxygen,  sulfur
                                     169

-------
dioxide, and C02 to form sulfates and carbonates.  The major elements in the
ash portion are sodium and inert oxides.
     Based on a combustion analysis, an F-factor for black liquor can be de-
veloped.  The F-factor (F.) for a typical sulfate pulping black liquor is
                                7?       '
approximately 51.07 dscf/lb BLS.    This value assumes stoichiometric con-
ditions and does not include inleakage or soot blowing.  The amount of water
vapor created by the combustion of hydrogen in the liquor can be calculated
by use of the same method, and is defined as F_.  The F  for a typical sulfate
                                              v5p
black liquor is approximately 7.99 scf/lb BLS.
     If the oxygen content and temperature are known, the flue gas volume can
be calculated from the boiler firing rate with the appropriate corrections
for the moisture in the flue gas.  Moisture in the flue gas results from
moisture in the fired liquor, moisture from the direct-contact evaporator,
and moisture added by soot blowing.
     If a more exact or plant-specific value is desired, the F-factor can be;
calculated from the stack test gas volume, the flue gas oxygen content, and
the firing rate of black liquor solids.
     Using this method, the operator, inspector, or environmental personnel
can determine, on a day-to-day basis, the flue gas volume being treated by
the ESP without the expense of using a Pitot tube to determine stack flue gas
volume.
     When flue gas oxygen increases above the normal ranges, the source of
inleakage should be identified immediately, and appropriate repairs should
be made to reduce the inleakage.  Failure to reduce the inleakage not only
results in excess emissions, but the cooling effect of the ambient air causes
reduced power input, excess sparking, and corrosion.
     For inspection purposes, the oxygen content of the flue gas stream can
be obtained by using portable instruments and by taking multiple readings
across the stack.  Care must be taken to avoid  inleakage stratifications that
may occur along the duct walls; this value would not be representative of the
major portion of the duct cross section.
     Corona Power--The power used by the ESP  in charging the dust particles
is measured in watts and is the product of secondary voltage and secondary
current.  The particulate removal efficiency  depends on the rate of particle
                                     170

-------
 charging  and  on  the electrical  field potential  in  which  the  particle  is placed.
 •It becomes  imperative  that the  maximum corona power  be maintained  in  each
 field  to  achieve the desired  performance.
     The  corona  power  in  a recovery boiler  ESP  increases  from  inlet to outlet
 as dust is  precipitated out of  the  gas stream.  Most units are  designed with
 an expected power consumption level  and an  expected  secondary current in each
 field.  The power consumption is  expressed  in watts  and  the  T-R set is sized
 accordingly.   Because  there is  a  loss  of energy in the transformer and in the
 rectification  process, the secondary-side power levels do not agree with the
 primary-side  power levels.  The conversion  efficiency ranges from  65  to 80
 percent depending on the  size and design of the T-R  set and  the ratio of T-R
 set operating  point (milliamps) to  T-R rating (milliamps).   The conversion
 efficiency  is  highest when the  T-R  set rating and operating  point are closely
 matched.73
     The  secondary current for  design  purposes  is expressed  in  milliamps per
 foot of electrode in each  field.  The  values generally are in the range of
 0.02 mA/ft  in  the inlet field and increase  in successive fields  up to about
 90 percent  of  the T-R set  rating.
     The  secondary voltage applied  to  the electrode  is maximized to provide a
 strong  field strength  (kilovolts) and  is limited by  electrode-to-plate clear-
 ance or T-R set  rating.
     The  limiting  level of corona power  input is determined by  the dust
 loading,  gas temperature,  moisture content, and wire/plate clearances.  The
amount  of power  delivered  is  typically expressed in  terms of the physical
size of the ESP  (watts per  square foot)  and is  based on the amount of flue
gas being treated  (watts/1000 actual cubic feet per minute).   The efficiency
of the  unit is directly related to the power delivered to the gas stream,
i.e., watts/1000 actual cubic feet per minute.
     Because many  studies  have indicated a strong relationship between corona
power and collection efficiency,  it is necessary to monitor and maintain a
high level of corona power.  The  inlet fields remove 90 percent of the par-
ticulates; therefore, maximum power levels must be maintained in the  inlet
fields.  The optimum power distribution  increases linearly from inlet to
outlet.
                                     171

-------
     To prevent corona power degradation due to misalignment,  dust cake
buildup, or T-R set failure, the boiler operator should record hourly T-R
meter readings on the boiler log sheet and note any deviations from normal
readings.
     Typically, ESP and boiler operating conditions are not recorded during
the stack test period.  Without these data, a comparative baseline (watts/1000
actual cubic feet per minute) cannot be established.  The comparative base-
line allows an accurate evaluation of the day-to-day operation.  Long-term.
degradation of corona power levels, which can occur because of the loss of
rapper effectiveness, increases in flue gas volume, misalignment, or changes
in T-R set controllers, is seldom noticed over a period of months, even
though the overall efficiency may be decreasing.  In most cases, immediate
short-term failures such as rapper control loss, T-R set controller failure,
wire breakage, drag chain failure, or insulator failure are indicated by
changes  in corona power levels, which can occur over a few minutes or several
hours.
     Daily review of  corona power levels by supervisory personnel and compari-
son with normal  values can  permit a rapid and  correct  diagnosis  of maintenance
problems before  they  result in  excess emissions or  catastrophic  failure  of
the unit.
     Most recovery boiler ESP's are designed with  two  parallel chambers, which
allows  internal  maintenance to  be performed on one  chamber while the  other
chamber carries  the gas volume.  Maintaining compliance with  emission limits,
however, requires  that  the  boiler load  be  reduced  to  keep  the total  gas
volume compatible  with  the  reduction  in collection  plate area because all  the
gas  is passing through  one  chamber.
      In those systems that  have a  single T-R set  per  field,  the  T-R set  is
 installed in a double half-wave design.  The T-R  set  controller  is typically
 designed to maximize  power  input based on secondary-side operating param-
 eters.   Power input to  the  unit is  limited by  the side with  the  lowest spark
 point (i.e., closest clearance, heaviest dust  cake buildup,  or a section with
 a cold air or oxygen  stratification in a single lane).  For  determining  if
 one chamber is limiting total power input, alternate  T-R taps should be
 grounded and the power, to each chamber evaluated.   Major deviations in voltage
                                      172

-------
 or  current  levels at  the T-R  set  limit or spark point  indicate clearance
 problems, rapper failures, or salt cake buildup on the plates.
     Gas Distribution—For optimum performance of an ESP, the velocity of
 gas passing through each gas  passage or lane must be approximately equal.
 According to the Industrial Gas Cleaning Institute (IGCI), the maximum
 deviation allowed across the  ESP  inlet face is +10 percent of the mean veloc-
 ity.  This  distribution of velocity is achieved by using inlet turning vanes
 and distribution plates.  Figure  3-72 shows a typical perforated gas distri-
 bution plate that has moved from  the vertical plane and become plugged.
 Erosion of  the vanes and plates or pluggage of the distribution plate causes
 a deviation in the gas distribution.  A serious deviation can reduce corona
 power because of stratification of the gas stream.
     The gas volume to each chamber is balanced and modified by inlet and
 outlet dampers.  Buildup on these dampers combined with pluggage of the dis-
 tribution plate can cause imbalance in chamber gas volume.  A Pitot tube
 should be used to make periodic checks of the gas volume to each chamber.
 Balancing of the gas volume based on damper position or static pressure drop
 can be very misleading.
     Drag Chains—Particulate collected on the plates and wires is removed by
 rappers and falls to the bottom of the ESP.   A drag chain assembly is used
 for continuous removal of the particulate cake from the bottom of the ESP via
 a discharge screw conveyor.  Failure to remove the salt cake can result in a
 buildup in the electrical field and a short in the T-R set, which can cause
 serious permanent misalignment of the wire frame assembly.
     Typical malfunctions of the drag chain system involve chain breakage,
 misalignment of the drags, sprocket failure, and/or motor failure.  Failures
 of  the chains and sprockets require isolation of the chamber to make the
 necessary internal  repairs.
     Two types of drag chains are used, longitudinal  and lateral.   The longi-
 tudinal drag-chain system moves across electrical  fields in the direction of
 gas flow.  Sneakage of the gas below the treatment zone between fields is
 prevented by baffles located between fields at a point above the drag chain.
The baffle does not completely isolate the field,  and some sneakage occurs
when the drag chains are in motion.   Failure of the baffles as a result of
                                     173

-------

             ••'••
          *1*5*
           '•!•*
                   •.;.•• -<-.»«^^i • -T:. , • "--g"**  ..• .. •••^r-,^j,--^^^:'j«
             -.:?•.•••• .-'?-•'- •••^•:-^/.'yf^,-sm^
             +  --/••''.•  -.-•-;•.  -r""-s^p

             >v':: V:/:^^:^^^&
             |:  ,.;!  •- ••: - ^VmV^fil*S
 i.-^;-"1^^ ^K '  • •'•V  ^F-p-  -• '- - ..  ' -.-".• •- . .:.'•-•;*..'-'• b-^^ '^- . •;• •:•....•.-:'. .:.•'•-•-..>-.>;;,. ;.:-.-f..^-.I'^y'-V?^^
 *';t^***Xv-^:
   (iS*^:^!^^:

  >   - >  •   ' r~T !P V£*

   ••••' ' '•"•-:'"'.-'''iS®



:;fe'^S^^
y.'' ;,, ''•T/^>VV'J^V-'''^^^0**W
       i»J-
            :•'• >^:
  &«<•,
Figure 3-72. Example of a plugged distribution plate.

                174
                                66

-------
corrosion or misalignment results in increased emissions due to sneakage
below the treatment zone.
     The lateral drag-chain system consists of an individual chain system in
each field that moves perpendicular to the gas flow.  Positive baffling is
provided between each field to prevent sneakage.
     Wet-Bottom ESP—In most recovery boiler ESP designs, salt cake rapped
from the plates is allowed to fall into a pool of black liquor maintained in
the bottom of the ESP.  Agitators are used to dissolve the particulate in the
liquor.  The black liquor that has been concentrated in the direct- or
indirect-contact evaporator is pumped into the ESP bottom, where it is en-
riched with the collected salt cake.  The agitation is slow and a crust of
salt cake is normally maintained on the liquor surface.  A continuous flow of
liquor is maintained in the bottom and is controlled by the boiler firing
rate.  The level is adjusted to maintain a positive seal under the antisneakage
baffles between fields.  If the agitation or mixing efficiency is poor, salt
cake may build up in the corners of the bottom and possibly cause bridging
and grounding of the electrical discharge system (i.e., clinker formation).
Because overfilling the bottom can ground the system, the liquor level must
be monitored continuously.  Corrosion of the baffles between the fields and
the shell wall is also a serious problem in wet-bottom ESP's.
     Rappers—Rapping is required in three areas in an ESP:  1) the collec-
tion plates, 2) the discharge wires, and 3) the distribution plates.
     Collected salt cake must be removed from the plates to maintain power
input levels (secondary current) .in each field.  Rapping effectiveness
depends on the salt cake properties, the temperature, and the transmittahce
of energy through the anvil/support system to the plates.  Loose or corroded
linkages dampen the vibration and result in ineffective removal of salt cake
in the lower portion of the plates.
     Salt cake must be^removed from the wire to reduce the effective wire    '
diameter and keep the corona initiation voltage low, which increases the
field strength and the power input.   Rapping of the wires is similar to plate
rapping, but fewer rappers are used per field.  Effective rapping is accom-
plished by creating a standing wave in the wire during the rapping process.
As in plate rapping, efficient transmittal of rapping forces is required to

                                      175

-------
remove the collected salt cake.  As a result, loose connections and corrosion
must be prevented to ensure effective rapping.
     Distribution plates must be rapped to remove deposits and prevent mal-
distribution of the gas flow.  Generally, a rigid connection is made between
the rapper shaft and the center of the perforated plate.
     Figure 3-73 shows a typical rapper pattern on a modern ESP using external
top rapping.  The number of.rappers per field and per chamber depends on
rapper size and plate design.  Usually there are more plate rappers than
discharge rappers.  The number of rappers is based on.the plate area and the
                                                        ^y
linear feet of discharge wire per field.
     Plate and wire frame rapping can be accomplished by magnetic impulse
gravity impact (MIGI) rappers, pneumatic cylinders, or external or internal
rotating hammers.
     Rapping frequency depends on dust loading and ESP design.  Typically,
the inlet fields are rapped more frequently than the outlet fields because
more dust is collected there.  Rapping intensity may be changed to remove
sticky or cohesive dust.  Overrapping can result in fracture of the dust cake
and resulting rapper reentrainment puffs or a snowing condition.
     Rapping intensity in MIGI rappers can be changed by increasing the
voltage to the integral DC coil (increasing lift) or by changing the weight
of the piston.  Generally, the piston weight is 20 pounds, and the lift
height can be adjusted to yield an impact of between 6 and 20 ft/lb.  The
amount of plate area per rapper is defined by the design.
     Typical failures of MIGI rappers occur as a result of coil failure
(i.e., open coil).
     Pneumatic impulse rappers are double-action air cylinders with integral
pistons.  At a set pressure, the frequency and impact force of the rapper are
defined. Various  kinds of rapper failures can cause the piston to freeze.
The most common failure results from water in the compressed air lines.
Water causes sludge or rust, which restricts the piston motion.  Condensation
as a result of air compression must be removed through  the use of traps,
dehydration units, or desiccant units.   Failure to install or maintain drying
equipment is the main cause of pneumatic rapper failure.
     Failure of rappers can also occur as a natural result of  piston  seal
wear.  A normal preventive maintenance program of replacing seals can mini-
mize this method  of failure.
                                      176

-------
X X
0 0
X X
X X
0 0
X X
X X
o o
X X
X X
0 0
X ' X
X X
0 0
X X
X X
0 0
X X
X X
0 0
X X
X X
0 0
X X
                             X            X
                                   X
Figure 3-73.  Typical rapper layout on a modern two-chamber precipitator.
                                  177

-------
      Pneumatic  rappers  are  activated  by  opening  an  air  valve  for a preset
 period  of time.  The  valve  is  opened  by  an  electrically activated solenoid.
 Although*  solenoid  failure is infrequent,  it can  cause rapper  failure.
      Vibrators  consist  of two  plates  separated by an air gap.  The upper
 plate is  attached  to  an electric  coil.   Passage  of  an alternating current
 through the  coil creates a  strong magnetic  field in the upper plate, which
 causes  the plates  to  come together with  great force.  Upon  impact, the elec-
 tric  circuit is broken  and  the gap is opened through the use  of compression
 springs.   The rapid closing and opening  of  the circuit  results in a high-
 frequency vibration in  the  rapper body,  which is attached to  the rapper rod.
 Rapping intensity  can be increased by changing the  voltage  to the vibrators.
 Failure of the rapper occurs as a result of spring  failure, a coil short, or
 a change  in  air gap setting.
      Falling-hammer rappers consist of a  number  of  swinging hammers attached
 to a  rotating shaft.  As the shaft rotates,  the  hammer  is brought into an
 elevated  position, from which  it  falls in an arc and strikes  an anvil attached
 to a  number  of plates or wire  frames.  The  force applied to the anvil is de-
 termined  by  the weight  of the  hammer, which can  be  changed  by increasing the
mass.   The frequency of the strike is determined by the rotation rate of the
 hammer  shaft.  Rapper failures occur as  a result of hammer/anvil misalignment,
 hammer  fracture, hammer/shaft  connection failure, and shaft chain-drive
 failure.   The frequency, duration, and pattern of rapper activation are de-
 termined  by  either electrical  or  mechanical  control systems.
      One  mechanical control system consists  of a synchronous  motor with an
electrical contact on the edge of a wheel.   Electrical  contacts located along
the perimeter of the wheel   are attached  to  individual rappers.  The pattern
of rapping can be changed by adjusting the  location of  the  leads.  The total
time  between rapping cycles can be adjusted  by changing the wheel rotation
rate.   Failure of the control system can occur if the contacts become loose,
burnt, or misaligned.
     A second type of mechanical   rapper control   system  consists of individual
contacts for each rapper, which are activated by a  cam.   A number of cam/
switches are attached to a   shaft with the cam lobes offset.    Rotation of the
cam shaft results in a preset rapper frequency and  pattern.    The cycle time
can be changed by changing   the rotation rate.  System failures result from
                                      178

-------
burning of cam switch contacts, cam breakage and wear, and synchronous motor
failure.
     Several types of solid-state rapper control systems are in use.  Most of
these systems allow changes in frequency, intensity, duration, and pattern of
individual rappers or group of rappers.  Failures of these rapper controls
result from thermal decomposition of solid-state components (resistors,
transistors, diodes, transformers,, etc.), loose wires and contacts, and
failure of breakers.
     Insulators—Insulators are used to isolate the high-voltage distribution
system from the ESP shell.  Cracking of the insulator as a result of moisture
or electrical tracking can cause a loss of a field (T-R section).
     Tracking occurs when dust or moisture accumulates on the insulator
surface.  The deposit provides a conductive layer through which the high
voltage may pass to the ground.  The circuit created generates heat on the
surface of the insulator that may eventually result in insulator cracking.
Depending on its location, insulator failure can cause a short that trips out
a T-R set.
     Insulator tracking may be, identified through use of an air load or gas
load test.  A typical pattern generated by insulator tracking is shown in
Figure 3-74.74
     Clearance Between Plates and Wires—The clearance between plates and wires
must be maintained at the design tolerances to prevent premature sparking and
to allow maximum power input.  Reduction in clearance can occur if the upper
or lower wire frame is not aligned with the plates.   The alignment is changed
by adjusting the standoff insulator position.   Clearance problems in wire
frame designs can occur as a result of- warpage of the wire frame or misalign-
ment of the wire frame alignment system.
     Plate-to-wire clearance can also be reduced as a result of plate warp-
age.  Minor warpage problems can be solved by physical straightening.  Severe
clearance problems can be corrected by removal of wires in the area of the
warpage.  Care must be taken to minimize the removal of wires within a single
passage or lane.
     Plate warpage can be caused by corrosion, unequal heating during startup,
fire in the ESP, or air inleakage.  Severe uncorrectable plate warpage that
limits corona power input may require replating of the unit.
                                      179

-------
 1000





   900






   800






   700





^  600


UJ



I  500

>-



I  400

CD
O
UJ
CO

   300






   200






   100






     0
__    A * SPARKOVER
                                             A
                                                   REFERENCE

                                                     CURVE
                                                  LEAKAGE

                                                 COMPONENT   —1
            10    15
                               20
25
30
35    40
45
50
                     SECONDARY VOLTAGE, kV
Figure 3-74.  Typical pattern generated by insulator tracking.



                              180
                                                           74

-------
     Corrosion—Corrosion appears to be the most serious maintenance problem
in the long-term operation of recovery boiler ESP's.  Corrosion attacks the
ESP shell and internal components.  Advanced corrosion is accelerated by air
inleakage through corroding areas in the ductwork, around access doors, or in
areas near the liquor-flue gas interface in wet-bottom units.  Corrosion in
internal areas causes a reduction in the removal of collected particulate
from plates and wires because rapper effectiveness is severely impaired.  A
survey by the Technical Association of the Pulp and Paper Industry (TAPPI) of
19 noncontact recovery boilers installed between 1974 and 1979 indicated that
63 percent had some corrosion problems and 26 percent had severe corrosion
         75
problems.    Based on the operating conditions of the 19 boilers, the average
temperature of those with serious corrosion problems was 361°F.   The average
temperature of those reporting no corrosion problems was 384°F.   Figure 3-75
shows an example of severe corrosion of collection plates in .an  ESP.
       Figure 3-75.   Example of severe corrosion of collection plates.
                                     181

-------
                                       76
     Corrosion  in a mild  steel  component  has  the appearance of multiple, thin
metal  layers  separated by areas of bright orange granular material.  As
oxidation occurs, the metal expands greatly in volume.  The loss of a few
thousandths of  an inch may expand to  1/2  inch as a result of crystal growth
and restructuring.  This  expansion, when  in a confined area such as between
plate  mounting  brackets or structural components, can cause a great deal of
stress and force components apart.
     Corrosion  occurs at  a faster rate  in the colder areas of the ESP.
Localized cooling occurs  when heat loss through the shell is highest, i.e.,
where  outside stiffeners  or structural  columns are attached to the shell.
     The primary corrosive agent in kraft recovery boiler,ESP's is sulfuric
acid.  Flue gases from the boiler contain hLO vapor with a high concentration
                                                                7fi
of SOg.  A portion of the S02 is converted to S03 in the boiler.    At tem-
peratures below 415°F, 99 percent of  the  S03 vapor combines with the water
present to form sulfuric  acid vapor (H2S04).  As the temperature of the gas
stream is reduced, the H2S04 vapor becomes saturated and forms an acid mist
     The temperature at which the sulfuric acid mist condenses on a cool
surface is defined as the acid  dewpoint.  The dewpoint actually is the begin-
ning of the saturation process, and the exact acid dewpoint temperature
cannot be determined.     The rate of  condensation reaches a maximum at 40° to
60°F below the  theoretical dewpoint.
     Severe corrosion occurs when water condenses on surfaces and dilute
acidic solutions are formed.  The rate  of corrosion is aggravated by air in-
leakage.
     The dewpoint temperature of uncombined water is different from the acid
dewpoint.  The  typical water dewpoint in  recovery boiler flue gas streams is
165°F, which must be avoided.
     The most severe corrosion  in wet-bottom ESP's occurs in the area above
the liquid level and below the  treatment  zone.  This area is baffled and is
not a  part of the main gas volume passing through the ESP.  Vapors from the
black  liquor in the bottom are  extremely  corrosive.  The activity of the
                                                                    78
vapors increases with high oxygen and sodium sulfide concentrations.    Also,
water  vapor raises the local dewpoint temperature.  The temperature of the
black  liquor, which is normally below 180°F, results in a cool shell tempera-
ture surrounding the liquor.  This cool shell  temperature causes a gradual
182

-------
79
 decrease in shell  temperature between the treatment zone and liquor level.
 The lower wall  temperature is usually below the acid dewpoint and near the
 moisture dewpoint.
      Maintenance of a uniformly high temperature in the ESP is important in
 reducing the rate of corrosion.  The temperature may be increased by reducing
 air inleakage,  insulating the shell, and heating the shell.
      The temperature within the ESP is not uniform, and variations in the
 shell  occur as  a result of contact with structural  members, degree of in-
 sulation,  exposure,  and orientation.  Areas of low  gas circulation in the
 ESP typically have  the lowest temperature and highest rate of corrosion.
      In  general, a  well-constructed, well-insulated steel  shell  ESP will
 experience minimum-corrosion at flue gas temperatures above 350°F.
 Units  with a flue gas temperature between 243°F and 265°F  have problems
 in  the areas of highest heat loss.   Units with temperatures below 300°F re-
 quire  supplemental  heating.
     The amount of  heating required varies with flue gas temperature,
 degree of insulation, and other environmental  factors such as wind loss
 and degree of exposure.   The heat requirements are  generally 500,000 Btu/h.
     Table 3-22 summarizes the parameters  and potential  malfunctions that
.affect emission rates.   As noted in the table, many parameters are inter-
 related  and may have more than one effect.   For this reason, the inspector
 should be able  to confirm as many operating parameters as  possible during the
 performance tests to define  deviations from normal  operation or practice.
 3.3.1.6   Inspection  of Recovery Boiler--
     This  subsection summarizes the activities associated  with the inspection
 of  kraft recovery boilers and ESP's.   It also identifies the kinds  of data
 that should be  collected during an  inspection  and the procedures that should
 be  used  to evaluate  these data.
     Both  plant personnel  and regulatory agencies make inspections  of process
 and control  equipment.   Although the thoroughness and scope of these inspec-
 tions  may  be vastly  different,  they have a common goal--to determine the  com-
 pliance  of the  source with applicable State and Federal  emission limits con-
 tained in  the SIP and the NSPS.
     To  achieve the  stated goal  of  determining the  compliance status of
 the source,  the inspection must be  well  planned,  and sufficient  time must be
                                      183

-------
     TABLE 3-22.
SUMMARY OF THE EFFECTS OF KEY RECOVERY BOILER OPERATING
               PARAMETERS
Parameter
            Factor
          influenced
       Effect
Firing rate
Char bed tempera-
ture
Primary air
Char bed height



Secondary air




Excess air
Firing residual
oil containing
high sulfur

(continued)
  Flue gas volume
  Vertical velocity in com-
  bustion zone
  Primary air volume
  Sodium and sulfur vapor
  pressure

  S02 content of flue gas


  Flue gas volume

  Vertical velocity in com-
  bustion zone

  Sodium and sulfur evapora-
  tion rate due to tempera-
  ture and diffusion condi-
  tions

  Char bed temperature


  Rate of primary air
  Vertical velocity in corn-
  bustiorf zone

  Excess air

  Vertical velocity in com-
  bustion zone

  Flue gas volume

  Increased S03 formation
  Increased SO3 formation
                                     184
Changes ESP efficiency

Changes uncontrolled inlet
grain loading

Changes diffusion condi-
tions and affects rate of
evaporation of sodium from
smel t

Changes uncontrolled par-
ticulate grain loading

Changes composition of dust
cake

Changes ESP efficiency

Changes uncontrolled par-
ti cul ate emission rate

Changes uncontrolled par-
ticulate emission rate
Changes uncontrolled par-
ti cul ate emission rate

Changes uncontrolled par-
ti cul ate emission rate and
flue gas volume to ESP

Changes uncontrolled par-
ticulate emission rate

Increases flue gas volume

Changes uncontrolled par-
ticulate emission rate

Changes ESP efficiency

Changes particulate com-
position; reduces ESP power
input and efficiency

Changes particulate com~
position; reduces ESP power
input and efficiency

-------
TABLE 3-22 (continued)
Parameter
          Factor
        influenced
                                                         Effect
Primary tempera-
ture
Black liquor
chloride content

Black liquor
heat value
 Sodium-sulfur
 ratio

 Flue gas  volume
 Corona power


 Corona initiation
 voltage

 Corona power dis-
 tribution within
 chamber
 Corona power per
 chamber
 Superficial
 velocity
 Flue gas oxygen
 at ESP inlet
 (continued)
Smelt bed temperature
Smelt bed elemental
equilibrium

Flue gas volume

Char bed temperature

Smelt bed elemental
equilibrium

Superficial velocity
Specific current density

Specific corona density
(W/1000 acfm)

Buildup of dust on elec-
trodes

Buildup of dust on elec-
trodes
Plate alignment

Buildup of dust on elec-
trodes
Plate alignment

Dust reentrainment during
rapping
Increased  total gas  volume

Flue gas volume
 Increased  S03 formation
                     Flue gas temperature
                                      185
Changes uncontrolled par-
ti cul ate emission rate
Changes TRS emission rate

Changes uncontrolled par-
ti cul ate emission rate

Changes uncontrolled par-
ti cul ate rate
Changes ESP efficiency

Changes uncontrolled par-
ti cul ate rate

Changes ESP efficiency
Changes ESP efficiency
 Changes  ESP efficiency
 Changes  ESP  efficiency
 Changes  ESP  efficiency
 Changes ESP efficiency
 Changes ESP efficiency

 Changes dust properties and
 increases plate deposits;
 reduces ESP power and effi-
 ciency
 Increases sparking, which
 reduces power and effi-
 ciency

-------
 TABLE  3-22  (continued)
Parameter
Flue gas tempera-
ture



Factor
influenced
Acid dewpoint
Moisture dewpoint



Effect
Corrosion
Structural failure
Increases air inleakage
Reduces corona power
Reduces ESP efficiency
 provided to acquire the necessary data on boiler and ESP operating conditions.
 This section outlines the overall scope of'the inspection and the procedures
 the inspector may use to determine the compliance status of the kraft recovery
 boiler with the particulate emission limits  defined by applicable State or
 Federal regulations.   The approach the inspector should take is to identify
 those operating conditions or variables that indicate operation outside the
 accepted norms for a  particular boiler/ESP system.   Normal  values or condi-
 tions are established during the initial  performance stack  test or are based
 on the accepted state-of-the-art.   This approach, which is  generally referred
 to as a "baseline approach" to source evaluation, may be used by both source
 personnel  and regulatory agencies  to initiate a  more detailed analysis or  to
 trigger a  performance stack test to verify compliance with  the emission stanr
 dard.
      The early identification  of operation and maintenance  (O&M)  problems
 reduces the  extent and  the occurrence of  excess  emissions and allows  the
 plant to schedule outages  or make  on-line adjustments  to maintain  production
 and operate  within the  prescribed  emission limits.
     Most  inspectors make  visible  emission observations  in  accordance  with
 EPA Method 9 and  record  the  production  rate of the boiler.  This  level  of  in-
 spection is  usually not  adequate for  determining compliance or evaluating the
maintenance procedures of  a  recovery  boiler.    The inspector must visually in-
 spect the ESP and  record pertinent operating variables for  both the boiler
and the  ESP.  Key  parameters also must  be measured to allow the inspector to
determine if the ESP is being operated within  design limits and if there is
sufficient reason to require documentation of  compliance through a stack test.
                                     186

-------
     The following is a summary of specific things that should be checked
during the inspection.
     Opacity—The inspector should conduct an EPA Method 9 observation of the
recovery boiler plume,,either at a point just before steam condensation (in
the case of a detached water vapor plume) or just after steam dissipation (if
the plume is attached).  The readings should be made at 15-second intervals
and averaged over a 6-minute period.  The total evaluation time should be at
least 30 minutes and preferably over one ESP rapper cycle (outlet field rap-
pers).  The 6-minute averages should be plotted to identify any cyclic pat-
tern.  An example of such a pattern is shown in Figure 3-76.
     Transmissometer Data--In units equipped with opacity monitors, the in-
spector should record  the current 6-minute average opacity and review the
previous 4 hours of monitor output to determine if a cyclic pattern is occur-
ring.  To ensure that  the output values are accurate, the inspector should
request the plant to place the monitor in the calibration mode with respect
to zero and span.  As  part of the initial monitor certification, the  inspector
should have data available on the recorder scale factors and  effective stack
diameter.  Average opacity readings from the Method 9 observation should be
compared with average  transmissometer readings for identical  periods.  A major
deviation between the  values may  indicate possible monitor  error.   It should
be noted that the manual and instrument methods are not equivalent.   In  sources
that have real-time monitor output, instantaneous opacity spikes  (rapper re-
entrainment) will generally be  included  in the 6-minute average  and the  value
will  generally be higher than that  obtained  by the manual method.   The in-
spector should note  the frequency and magnitude of rapper spikes and  determine
if a pattern  is  occurring  (inlet  to outlet field  rapper pattern).   The opacity
and  rapper  reentrainment pattern  in each  chamber  should be  evaluated  if
separate monitors are installed in  each  duct.  Figure  3-77  shows  a  typical
monitor output with  severe  rapping  reentrainment  losses.  The opacity should
be compared with a typical  baseline value for  the boiler  during  known emis-
sion periods  (i.e.,  performance tests).   The opacity  data should be used to
evaluate  conditions  in the  ESP.
      Serious  deviations  in  opacity  between  chambers  can  indicate gas  flow
maldistribution, an  increase  in penetration  through  one  chamber  as  a  result
 of rapper failure,  inleakage,  or low power input.

                                      187

-------
                                       88T
                                     AVERAGE OPACITY,  %
                                       ro
                                       o
                                           CO
                                           o
en
o
 CQ
 c
 CO
 rn

 1
 -a

 a>

 o
 -h
 3
 C
 rt-
 n>

 Q)

 n>

 cu
U3
 fD

 o
•o
 o>
 o
 —1.

4?

•a
 EU
(D
-s

•
  CO
  o
t-i  CO


3>

m
en
m  f*


m
    tn
   (Tl

-------
00
                   Figure 3-77.  Typical  opacity monitor output with- severe rapping reentrainment losses,

-------
      Boiler Operating Conditions—The inspector should record such boiler
 operating conditions as steam flow (103 pounds per hour), flue gas oxygen
 (%), and flue gas temperature (°F) at the time of the inspection.   By com-
 paring these current operating conditions with the historic baseline obtained
 during the performance test, the inspector can determine if the boiler is
 operating at normal  production levels.   Major deviations from normal values
 should be evaluated  with respect to their impact on ESP performance and TRS
 and particulate emission levels.  Operating conditions, continuous emission
 monitors (TRS and opacity), and ESP operating conditions should be used to
 determine if the boiler is in compliance with applicable standards.
      Boiler operating data that should  be measured during a performance test
 or an inspection are plant-specific.  Each boiler is  usually custom-designed
 and erected with a unique instrument and control  system package.   The level
 of instrumentation is specified by the  design engineer and purchaser (plant
 engineering department).   Based on the  size of the boiler,  its  cost,  and  the
 experience of the purchaser, the instrument package may range from a straight-
 forward package to one that is  very complex.   In  general,  a minimum  amount of
 instrumentation is necessary for safe operation of the boiler,  and all  facil-
 ities will  have this  level  of instrumentation.  More  complex instrument
 packages  can  include  an  automated computer control  system  that  allows  the
 source to optimize combustion and increase the overall  efficiency  of the
 operation.
      Most critical boiler  parameters  are  recorded  on  continuous strip  charts,
 or  circular chart recorders,  and  copies may  be obtained after the  stack test
 (at the end of  the day)  to  provide  the necessary documentation.  Most  mills
 require the boiler operator to record key  parameters  at set  intervals  on a
 log sheet or  in a  log book.   The  log  sheet is  typically divided into  the fol-
 lowing general measurement  areas:   black  liquor, auxiliary  fuels,  forced air,
 furnace drafts, gas temperatures, feedwater,  steam, and miscellaneous  items.,
     Table 3-23 lists the items or conditions  that must be  recorded during
 the  stack test or  inspection.  The  list is based on a  typical boiler and
would require adjustment for  individual  installations.
     Integrator readings also should be recorded at the beginning and end
of each test run for the following parameters:
                                     190

-------
      TABLE 3-23.  RECOVERY BOILER OPERATING PARAMETERS TO BE RECORDED
                  DURING PERFORMANCE TESTS OR INSPECTIONS
Parameter
Black liquor









Auxiliary fuels




Forced air









Variable
Liquor flow
Black liquor pressure
Gun size
Number of guns
Black liquor temperature
BLS to guns
BL flow to ESPa
BLS to ESPa
BL flow to evaporator3
BLS to evaporator9 .
Oil flow
Number of guns
Oil pressure
Oil temperature
Natural gas rate
Primary air flow
Primary air pressure
Primary air temperature
Secondary air flow
Secondary air pressure
Secondary air temperature
Tertiary air flow
Tertiary air pressure
Tertiary air temperature
Total air flow
Units
gal /mi n
(103 Ib/h)
psig
None
None
°F
%
gal /mi n
. (103 Ib/h)
V
la
gal /mi n
(103 Ib/h)
%
gal/h
None
psig
°F
103 fts/h
scf/min
in. H20
°F
scf/min
in. H20
°F
scf/min
psig
°F
scf/min
(continued)
                                    191

-------
TABLE 3-23 (continued)
Parameter
Furnace drafts





Gas temperatures





Feedwater


Steam


Chemicals
Miscellaneous

Variable
Furnace
Superheater outlet
Boiler outlet
Economizer outlet
ID fan inlet
Preci pita tor inlet
Superheater outlet
Boiler outlet
Economizer outlet
Evaporator outlet
ID fan outlet
ESP inlet
Flow
Pressure
Temperature
Flow
Drum pressure
Superheater temperature
Salt cake makeup
Flue gas oxygen (boiler outlet)
Black liquor heat value
Units
in. H20
in. H20
in. H20
in. H20
in. H20
in. H20
°F
°F
°F
°F
°F
°F
(103 Ib/h)
psig
°F
(103 Ib/h)
psig
°F
Ib/min
%
Btu/lb BLS
 To be used with correction factors to calculate BLS to guns  where not
 measured directly.
                                     192

-------
     o    Black liquor flow (10  pounds  or gallons)
     o    Steam flow (pounds)
     o    Steam used in soot blowing (pounds)
     o    Oil  flow (pounds or gallons)
     o    Natural  gas flow (103 cubic feet).
     Based on data obtained from the log,  steam tables,  integrator  readings,
and boiler design  data, the following values  should  be calculated:
     o    Average  steam flow for each run
                              3
     o    Average  BL fired (10  pounds/hour)  for each test run
                               o
     o    Average  BLS fired (10  pounds/hour)  for each test run
     o    Heat input (10  Btu/hour)  to the boiler for each test
          run for  each fuel fired (BL,  oil, natural  gas)
     o    Average  boiler output (10   Btu/hour) for each  test run
     o    Boiler thermal efficiency  (heat  output/heat input)
          for each test run
     o    Boiler excess air (percent)
     o    Pounds of steam per pound  of BLS fired.
     ESP Power Levels—The inspector should record ESP power levels (primary
current, primary voltage, secondary  current,  secondary voltage) for all  ESP
fields and chambers.  The spark rate should be estimated from a manual  count-
ing of meter deflections.  The inspector also should plot ESP functions by
field (inlet to outlet) for each chamber.   Deviations from optimum  values
determined from baseline or normal  values  should be  used to determine internal
ESP conditions and to analyze potential  emission levels.   If recent V-I
curves are not available, the inspector should request the plant  environ-
mental engineer or electrician to produce  a V-I curve for each field.  Data
from the V-I curves should be used to target  the inspection of the  rappers,
the gas distribution system, and local  cooling and to check for inleakage.
Serious deviations from normal values should  be evaluated with respect to
their impact on potential emission levels. Opacity  data for each chamber are
helpful in determining the effect of corona power levels.  In general, the
efficiency of the ESP follows the pattern  of  the Deutsch-Anderson equation or
Matts-Ohnfeldt equation for prediction of  emissions  as a function of corona
power and gas flow rates.
                                    193

-------
     ESP Rappers—The  inspector should determine if discharge electrode,
collection plate, and  distribution plate rappers are functioning.  Particular
attention should be given to areas identified as having decreased corona
power or areas where higher corona initiation voltages were noted.  Initial
evaluation of the rappers should be based on sound intensity and/or failure
to activate.
     Flue Gas Volume—Because ESP performance is affected by total gas volume,
the inspector should make an estimate of the volume based on black liquor
firing rate, flue gas  oxygen, and temperature.  Most plants monitor flue gas
oxygen at the economizer outlet rather than at the ESP outlet.  An estimate
of the flue gas volume must be based on ESP outlet conditions.  The inspector
should be equipped with portable temperature measurement equipment (i.e.,
thermometer or thermocouple) and portable oxygen measurement equipment (i.e.,
Fyrite oxygen analyzer).  The flue gas volume may be calculated from a plant-
specific F-factor (dry standard cubic feet/pound BLS) with correction for
flue gas oxygen5 moisture, and temperature..  Temperature and oxygen measure-
ments should be made at the outlet of each chamber where possible (accessible).
     The following presents the method used to calculate the flue gas volume
at the ESP inlet or outlet.  It should be noted that the corrections for the
flue gas oxygen are for dry standard gas volume not wet gas volume.
     Q =
          BLS
20.9
                 dry 120.9 - %02
460) °R
                                528°F
where
     BLS = black liquor solids firing rate to the boiler.
    Fd   » F-factor for black liquor solids in dscf/lb BLS.
    % Og s oxygen content of flue gas at ESP inlet in percent.
      F  = standard cubic feet of water vapor generated from combustion
           of hydrogen per pound of black liquor solids.
      FE = standard cubic feet of water vapor evaporated in direct
           contact evaporator.
      F  = standard cubic feet of water vapor added to flue gas stream
           as a result of soot blowing.
                                      194

-------
      TS =  temperature of the flue gas at the ESP  inlet in °F.
     The amount of water evaporated  in a direct-contact evaporator may be
determined  by using  liquor flow rates and liquor solids content entering and
leaving the unit  (see Figure 3-78).  This method is a simple mass balance
based on the assumption that the total amount of solids does not change in
the evaporator.   This is not strictly true because the liquor does adsorb
salt cake from the flue gas stream.  This effect is considered negligible,
however, in calculation of the water lost.  If a more exact estimate is
desired, adsorption  rates may be estimated depending on flue gas volume (actual
cubic feet/minute) and the uncontrolled boiler dust loading (grains/actual
cubic feet  per minute).  In general, a cascade type evaporator may remove 50
percent of  the uncontrolled particulate.  In most units this will increase
the total liquor  solids mass by less than 5 percent and usually will result
in less than a 2  percent error in the true gas stream moisture.
     If a more exact or plant-specific value is desired, the F-factor can be
calculated  from stack test gas volume (dry standard cubic feet/minute), flue
gas oxygen  content, and firing rate of black liquor solids.
     Using  this method, the operator, inspector, or plant environmental per-
sonnel can make a day-to-day determination of the flue gas volume being
treated by  the ESP without the expense of conducting stack flue gas volume
determinations with a Pitot tube.
     When flue gas oxygen increases above the normal ranges, the source of
inleakage should be identified immediately and appropriate repairs made to
reduce the  inleakage.  Failure to reduce the inleakage will not only cause
excess emissions, but because of the cooling effect of the ambient air, "will
also cause low-power input, excess sparking, and corrosion.
     The amount of steam used in soot blowing is not generally measured, but
based on discussions with ESP and boiler manufacturers and limited data from
pulp mills, the value is estimated to be 8 to 10 percent of the rated steam
                   81
flow of the boiler.     The value is expressed in pounds of water vapor per
minute,  which must be converted to standard cubic feet per minute.  The values
of F  and F .   may be compared with the values obtained during a stack test
as a check on the validity of the derivation.   The variables affecting these
values are too numerous to list here, but they include wood species and mix,
process  step variables, quantity of inorganic salt cake recycled to the

                                     195

-------
                  FLUE  GAS  (IN)
                                      FLUE GAS (OUT)
f
IQUOR (OUT)
EVAPORATOR
BLACK LI QUO
where
     A =
     i s
     B *
           water evaporated (Ib/min) = A p. - B p

gallons of black liquor to the evaporator
density of black liquor into the evaporator in Ib/gal
gallons of black liquor from the evaporator
density of black liquor out of the evaporator in Ib/gal
                      % BLS.
                A P.. (	TT^TT-) = B
                                               % BLS.
where
% BLS.. = solids content of black liquor entering evaporator
% BLSQ = solids content of black liquor leaving evaporator
Figure 3-78.  Method of calculating additional moisture in the flue gas
             stream due to direct-contact  evaporator.
                                196

-------
recovery boiler, percent solids in the black liquor, and heating value of the
black liquor.
     When deriving the black liquor F-f actor from stack tests or by theoretical
equations, it is convenient to work in terms of standard cubic feet of gas be-
cause it allows for addition of values without constant correction for different
gas conditions.  The ESP, however, must be analyzed at the actual gas condi-
tions (i.e. , at the measured temperature and oxygen content).  Once established,
the values of F.   and FtQ+ai tend to remain relatively constant provided no
significant changes in the process occur.
     The use of an established F-factor to determine gas flow through the ESP
requires relatively little calculation.  Only the following are needed:  value
of the F-factor (dry), firing rate of the black liquor, percent BLS, density
of the black liquor, and the temperature and oxygen content of the flue gas.
     The ESP dimensions can be obtained from engineering drawings (blueprints).
Using these dimensions will usually produce superficial velocity values that
are slightly lower than actual values.  The area input into the calculations
does not account for the cross-sectional area blocked by the plates and wires.
The calculated value should be in the range of 2.5 to. 4.0 ft/s, and the lower
values generally are recommended.  Obviously, as the superficial velocity in
an ESP decreases, treatment time will increase.  Also, if the superfical
velocity exceeds 8 ft/s, not only will the treatment time drop, but reentrain-
ment of captured particulate may occur as a result of the high velocity strip-
ping material off the ESP plate.  Thus, it is important to consider the gas
volume through the ESP.  Gas volume is especially critical if there is a pos-
sibility of high excess air levels resulting from air inleakage or improper
boiler operation, or if high gas volumes could occur from overfiring the
recovery boiler.
     Another value that should be checked is the actual SCA.  This value re-
                                                          o
lates the total available plate area to the gas volume (ft / 1000 acfm), and
when compared with design or baseline values, indicates ESP performance
capabilities.  Generally, an increase in the SCA (actual) means improved
performance, but other factors are involved; therefore, a comparison of actual
SCA with design or baseline values is not meaningful by itself.
     ESP Corona Power— The evaluation of ESP performance is based primarily
on T-R electrical readings from both primary and secondary meters.  The
                                     197

-------
inspector can define most of the problem areas and determine the ESP performance
level externally through the use of a combination of calculations and the
evaluation of trends present in the data.
     Secondary meter readings are preferred for an evaluation of ESP perform-
ance and diagnosis of ESP problems because they tend to reflect more closely
the power input characteristics of the ESP.  For example, the secondary
voltage at which secondary current is initiated can be used to determine if
ash is building up on the wires and if insulators are tracking.  These data
allow adjustment of the vibrator intensities for optimum power input and
indicate the need for cleaning or replacing insulators.  The secondary cur-
rent/voltage curve also can be used to diagnose clearance problems and indi-
cate plate warpage and/or wire frame misalignment on a section-by-section
basis.
     No voltage on the secondary side may indicate an open primary circuit.
The circuit breaker may be open or tripped or a reactor secondary may be
open.  High voltage on the primary side and no voltage on the secondary side
may be due to a faulty, open, or disconnected ESP; an open bus; or a faulty
rectifier.  Low voltage on the secondary side coupled with low voltage on the
primary side could be the result of leaks in the high voltage insulation,
buildup of dust in the discharge system, excessive dust on electrodes, or
                    CO
swinging electrodes.
     No secondary current and no secondary voltage indicate an open primary
circuit.  Irregular secondary current coupled with low secondary voltage
indicates a high-resistance short in the circuit, possibly due to excessive
               fi°
dust or arcing.
     Because primary meters can provide an-indication of the voltage and
current going to the ESP, they can be used if they are the only meters
available to obtain a more indirect measure of ESP performance.  For example,
no voltage on the primary side may indicate an open primary circuit.  The
circuit breaker may be open or tripped, or the reactor secondary may be open.
High primary voltage may indicate an open transformer primary or improper
connection of an ESP, a faulty, open, or disconnected ESP, an open bus, or a
faulty rectifier.  Low primary voltage may indicate leaks in the high-voltage
                                                                 en
insulation, excessive dust on electrodes, or swinging electrodes.
                                     198

-------
     An ammeter reading that shows no primary current  associated with no
primary voltage indicates an open primary circuit,  which may  be due to open
circuit breakers or an open reactor secondary.   Irregular  primary current
coupled with low primary voltage indicates a high-resistance  short in the
circuit.  The' causes of this condition include electrode short, excessive
dust on collecting surfaces, excessive dust on electrodes,  support insulator
                                                      CO
arcing, and the possible presence of foreign material.
     When recording and evaluating ESP meter readings,  the  inspector should
realize that trends in the voltage and current levels  are  the first indicators
of ESP performance levels.  Specifically, a gradual  rise in the ESP secondary
current levels should be apparent as one progresses from the  inlet to the
outlet fields (Figures 3-79 and 3-80).  The secondary  current level seldom
exceeds 250 milliamperes (mA) in the inlet fields.   Typical designs call for
a current level of 0.02 mA per linear foot of wire  length  in  the inlet fields.
Sparking in the inlet fields is evidenced by deflection of the meters during
a spark.  When the meters indicate progression toward  the  outlet field,
secondary current levels should increase and sparking  should  decrease, with
almost no sparking occurring at the outlet.  Most ESP  outlet  field T-R cur-
rent levels should be at least 85 percent of the T-R set current rating
(e.g., if the secondary current rating was 1000 mA  on  the  outlet T-R, a
design reading of at least 850 mA is normally expected) and in the range of
0.06 to 0.08 mA per linear foot of discharge wire.
     Another trend the inspector, may note is a gradual  decrease in the secon-
dary voltage from inlet to outlet.  Larger ESP's (with five fields or more)
often experience lower voltages in the inlet field  due to  sparking, an in-
crease in voltage in the middle fields; and then a  slight  decrease in voltage
in the outlet fields.
     The gas entering the inlet field of the ESP contains  the greatest
concentration of particulate, and the greatest quantity of particle charging
occurs in this field.  Consequently, a great many electrons are captured
during charging, and the rate of charge transfer from  the  discharge to the
collection electrodes is lessened because the mass  of  the  particles migrating
toward the plates is considerably larger than that  of  the  electrons or
charged molecules (ion mobility).  In addition, more force  is required to
                                    199

-------
LU
CC
O
§
t/1
                                         J_
                             2            3

                            FIELD NUMBER
                                                                     0.10
                                                                    0.08
                                                                    0.06
                                                                    0.04  >•
                                                                         , cc.
                                                                          C£.
U



QC
                                                                          o
                                                                          <_)
                                                                          LU
                                                                          tn
                                                                    0.02
Figure 3-79.  Optimum secondary  current distribution in ESP serving kraft
  recovery boiler, assuming uniform  rapping  and wire size in all fields.
                                 200

-------
ee.
 §
C/5
                                                      >CHAMBER B
                                                      •CHAMBER A
                                FIELD
Figure 3-80.   Secondary current pattern  for  two  ESP  chambers; Chamber A
        is having maintenance problems that  limit  power  input.
                                  201

-------
move the electrons from the corona discharge to the collecting  plates  because
the charged particles in the interelectrode space act as a large electro-negative
cloud that repels the electrons.  This is known as the "space charge"  effect
and is usually present only in the first and sometimes the second electrical
field.  The greatest sparking occurs in this field because this large  cloud
of charged particles tends to form numerous paths for interelectrode gas
breakdown (spark formation).  Thus, operation in this field tends to be a
balance between very high electrical field strengths and reduction of  spark
rates to moderate levels without excessive waste of electrical power.
     As the gas moves through the ESP and the particles migrate to the plate,
fewer particles remain to be captured and to inhibit the flow of electrons.
Therefore, less force (voltage) is required to obtain a high current flow.
Usually the increase in current is much greater  than the decrease in voltage,
and  the net effect  is an  increase in  power  input from ESP  inlet to outlet.
     The  relationship or  ratio  between voltage and current levels for each
T-R  is not constant throughout  the ESP (from inlet to outlet).  Changes  in
 the  voltage-current relationship  are  due  to capacitance and  resistance char-
 acteristics of the  particulate, as well  as  to  inefficiencies in  the circuitry
 of the T-R  set at various operating levels.  The relationship  between  the
 primary  and  secondary voltage and current levels will not be constant  from
 inlet to outlet for the same reasons. Thus, general  relationships  between
 primary  and secondary meter readings  are difficult to establish and will
 change with dust characteristics.  Over  very narrow operating ranges  within a
 T-R, however, the relationship  between primary and secondary voltage  and
 current levels may be considered linear.  This relationship  is useful  in
 evaluating ESP performance if a secondary meter is out of service and the
 corresponding primary meter on the other side of the transformer is operating.
      Although trends in the voltage and current levels  are important  in
 evaluating ESP performance, the corona power input provides  one of the most
 useful indicators.  Secondary meters are preferred in determining corona
 power because they are more indicative of actual power input to the ESP.
 Primary meters may be used for this calculation, however, if they are the
 only meters available, provided an appropriate  efficiency factor is used.
 Corona power calculations are  simply the product of the secondary voltage
                                      202

-------
multiplied by the secondary current to yield watts of power to the ESP field
from  the T-R.  This calculation should be made for each field of the ESP.
                                                                  *
When  both primary and secondary meters are available, the primary voltage and
current levels should be multiplied to yield primary power Input to the T-R
in watts.  The value produced by the primary meter must be higher than that
obtained from the secondary meter.  If the secondary power product (corona
power) is higher than the primary power input, then the values on the meters
are incorrect.  Isolating the malfunctioning meter, however, may be very
difficult.
      Electrical losses occur in the T-R during increases in voltage and rec-
tification to an unfiltered DC wave form.  In addition, losses in the T-R
control circuitry reduce the efficiency of transferring from primary input
power to secondary corona power.  This efficiency factor for the T-R may be
calculated through the use of the ratio of secondary power to primary power:
                    T-R efficiency = secondary power
                                 J    primary power
(Eq.  1)
     This value usually ranges from 0.55 to 0.85, although values up to 0.90
are occasionally observed.  In general, the value of the T-R efficiency in-
creases as the T-R approaches its rated output current level.  Thus, T-R
efficiencies tend to be lower in the inlet fields (0.55 to 0.60) and higher
in the outlet fields (0.80 to 0.85) because of limitations imposed on the
electrical operating characteristic by particulate load and space charge
effect.  A value of 0.70 to 0.75 is usually appropriate as an average for all
fields in the ESP.
     Another item that should be checked in multichambered ESP's with T-R's
for each chamber is a balance of power across the ESP.   The secondary current
level and power input for each field should be approximately equal across the
ESP.  Some relatively small differences may occur because of rapper sequenc-
ing, slight gas flow imbalances, or differences in internal alignment.   These
differences should not be large in most ESP designs, however, as chambers
with equal gas flow and equal  power levels give the best ESP performance.
     Once the power levels have been calculated and the patterns checked as
previously discussed, the corona power from each T-R should be added and
totalled to yield total corona power to the ESP (in watts).  When multi -
chambered ESP's are installed  with T-R's serving each individual chamber,
                                     203

-------
the corona power levels should be totalled for each  chamber  (to indicate
balanced power), and an overall total  corona power level  should be calcu-
lated for all chambers.
     Baseline test values, if available, should be compared  with  the actual
secondary current'and corona power levels.  If the gas volume through  the  ESP
is nearly identical to the test value and the meter  readings are  also  nearly
the same, then ESP performance is usually similar to that observed during  the
stack test.  It is usually not possible to compare design power levels to
performance levels, as these are usually not included in most design  informa-
tion.  A  key indicator of ESP performance is the specific corona power, which
provides  a useful value for determining whether ESP performance has  changed
significantly.
     Specific corona power is  calculated  by the following equation:

                                  total corona power  (watts)
           Specific  corona power  = total gas volume  (1000 acfm)

This value may  be calculated  for the  entire ESP or  for individual chambers.
The volume obtained by the modified F-factor  (divided by  1000) is substituted
 into this equation.  In general, the  higher the value of  the specific corona
 power, the higher the ESP removal efficiency.  Thus,  one  may determine whether
 ESP performance would be expected to increase or  decrease by evaluating the
 specific corona power.
      In general, the specific corona power needed to  meet NSPS for kraft  re-
 covery boilers is usually above 400 watts per 1000  acfm,  although acceptable
 performance may be obtained with lower specific corona power values  if there
 are no major problems with power distribution, inleakage, or rapper  operation.
 Equation 2 indicates that a decrease in gas flow through the ESP may lead to
 an increase in performance (constant corona power is assumed).  In  this case,
 however, the corona power is generally not constant, but increases  with de-
 creasing gas velocity  (volume)  because the bulk of the particles do not
 penetrate far enough  into the ESP  to inhibit power input (the opposite occurs
 with an  increase in the gas volume).  Thus, a decrease in gas volume may
 substantially  increase ESP performance.  This relationship  provides an impor-
 tant  reason for  evaluating firing  rates  and excess air levels.  It should be
 noted, however,  that  poor performance  is possible  (but less likely) with

                                      204

-------
specific corona power levels of more than 750 watts/1000 acfm and that specific
corona power provides a useful but not a sole indicator of ESP performance.
     Another useful indicator of ESP performance is power density (watts/ft2
plate area).  The value for the power density may be calculated by the equation:

                                                                     (Eq. 3)
Power density = corona power input
                 ft  plate area
     This calculation may be performed for each field by substitution of the
T-R corona power and the plate area associated with that T-R into the equa-
tion.  The increasing trend from inlet to outlet that should be evident may
range from 0.25 W/ft2 at the inlet to 5 W/ft2 in the outlet field.  This
calculation is useful for ESP's that do not employ equally sized fields.
The trend of secondary current and corona power is not smooth, but this cal-
culation will help "normalize" the values.  A corresponding calculation may
be performed on the secondary current (current density, mA/ft2) to normalize
the data.
     The substitution of total corona power and total plate area into the
equation will usually determine the overall power density.  Values of 1 to
3.0 W/ft  are typical and usually indicate good performance.  As with specific
corona power, however, performance may be poor even though the power density
may be high.
     A baseline test can be used to calculate ESP efficiency.  The baseline
test is used to calculate total corona power, gas volume, and efficiency or
penetration (1 - efficiency).  The values from the baseline test can be used
along with a modified version of the Deutsch-Anderson equation to calculate
a constant that may be used in subsequent calculations:68
                             p  = e-0.06 K(Pc/Q)
                              U
                                                 (Eq.  4)
where
          Pt = penetration
           K = constant
          PC = corona power, watts
           Q = gas flow, 1000's acfm
                                     205

-------
     This equation relates corona power input and gas flow to ESP  performance.
The original form of the Deutsch-Anderson equation assumed maximum and  non-
varying field density and related particle migration rate to the SCA of the   ,
ESP.  The particle migration rate, however, was a function of corona power
input; thus, corona power changes the particle migration rate and  the ESP
efficiency.  This equation is useful in predicting changes in efficiency
provided that wide variations in the specific corona power have not occurred
or the power distribution within the ESP is not seriously altered.
     The value of "k" will typically fall in the 0.1 to 0.25 range for  most
kraft recovery boilers.  Once the value of the constant has been calculated,
it can be used in subsequent calculations in which inspection data are  used.
Because the equation usually will overpredict performance within narrow
operating ranges, the efficiency predicted is based on the assumption that
all the power is used to collect particulate.  As substantial power decreases
occur, however, the equation may substantially underpredict performance
because the equation becomes very sensitive to variations in the specific
corona power.  In fact, the value of the constant begins to change with sub-
stantial increases or decreases of specific corona power, and the  predicted
performance may be incorrect by as much as half an order of magnitude when
applied over a very wide range of specific corona power.  The value of  the
constant and the behavior of the ESP will also change with inoperative  T-R's
(particularly those in the inlet fields).  The equation will, however,  pro-
vide a gross indication of a performance shift.
     The underlying limitation to the equation is the assumption that the
power input and the changes in power input are the only factors that affect
ESP performance once the SCA and gas volume are fixed, and that the value  of
the constant and the effective migration velocity are linearly related.
Actually, the relationship in the value of the exponent is probably a power
                                                ••
function similar to that presented in the Matts-Ohnfeldt equation, which
makes the equation less sensitive to variations in power input. Using  an
equation at conditions different from baseline values still requires care,
however.
     In summary, evaluation of ESP data includes plotting of secondary  current
to discern an increase from inlet to outlet; calculating corona power from
                                    206

-------
 primary  or  secondary meters;  evaluating  the  balance  in  the readings between
 parallel  fields;  calculating  total  corona power, power  density, and specific
 corona power; and comparing these values to  baseline readings.  If conditions
 are nearly  identical,  performance is  likely  to be similar.  For small changes
 of specific corona power, an  indication  of the magnitude of any performance
 shift may be estimated by use of a  modified  form of  the Deutsch-Anderson
 equation.   Care must be exercised not to use this equation when gross changes
 in ESP operation  have  occurred unless the values are modified appropriately.
     ESP  External Conditions—The inspector  should visually inspect the ESP
 shell to  determine the condition of the  insulation and  note any major points
 of oxygen inleakage and corrosion.  The  external inspection should be based
 on the analyses of flue gas oxygen  and ESP power levels.  Table 3-24 summarizes
 the parameters that the inspector should measure as  part of an ESP inspection.

  TABLE 3-24.  PARAMETERS TO  BE MEASURED BY  THE INSPECTOR DURING LEVEL III
                       INSPECTION OF RECOVERY BOILER  ESP
Parameter
Flue gas oxygen
Flue gas temperature
Primary voltage
Primary voltage
Secondary voltage
Secondary voltage
Location
ESP inlet
ESP outlet
ESP inlet
ESP outlet
Each field
' Each field
Each field
Each field
Units
01 .
to
of
h
nF
°F
Volts
Amperes
. Kilovolts
Mi li amperes
     Record RevJew--The inspector should review the records maintained by the
plant environmental engineer with regard to component failures, outages, and
trends to determine the performance of the unit between inspections.  Typical
records that should be reviewed include weekly inspection reports, quarterly
inspection reports, annual internal inspection reports, air-load and gas-load
V-I curves, work orders, wire breakage location charts, and the number of
component failures (e.g., rappers, insulators, wires).  To interpret the
                                      207

-------
maintenance records adequately, the inspector should request a copy of the
ESP specifications and ESP construction blueprints (arrangement, wire frame,
plate detail, etc.).
3.3.2  Smelt Dissolving Tank
     The following subsections describe the smelt dissolving process, identify
the major emission sources, discuss the control techniques for minimizing
emissions from the smelt dissolving tank, discuss the possible malfunctions
associated with the operation of the smelt dissolving tank, and present the
inspection procedures for the smelt dissolving tank and associated control
equipment.
     The particulate emissions from the smelt dissolving tank are generally
controlled with low-energy scrubbers.  The smelt dissolving tank malfunctions
that affect emissions are generally associated with the various types of
scrubbers that are used to control the particulate emissions.  Table 3-25 on
page 215 summarizes the malfunctions and identifies their potential impact.
     The inspection of the smelt dissolving tank area usually involves a Level
II Inspection because of the limited accessibility of sampling locations.  The
level of inspection depends a great deal on the availability of instrumenta-
tion and access to the control equipment.  A detailed discussion of the items
to be covered in Level I, II, and II inspections for smelt dissolving tanks
is presented in Section 3.3.2.4.
3.3.2.1  Process Description--
     Molten smelt composed of sodium sulfide and sodium carbonate  is drained
from the recovery furnace hearth through smelt spouts.  The smelt  is dis-
charged into a water-filled vessel referred to as the dissolving tank.
"Green liquor" is formed by the incorporation of the smelt into water through
quenching.
     The melting point of the smelt  is defined by the composition,  i.e.,  per-
cent Na2C03 and Na2S.  Figure 3-81 presents an equilibrium diagram for the
system.12*82  The hearth is generally operated at a high temperature to ensure
free-flowing smelt.  Temperatures in the smelt spout are typically between
1600° and 2000°F.
                                      208

-------
              LJ
              QC
              UJ
              o.
                                                         2I56°F
                                                       	I380°F
     1300
No-CO, % 100    80   60    40
         0     2O   4O    GO
                                                 20
                                                 80
                       COMPOSITION OF SMELT, % BY WEIGHT

      Figure 3-81.  Equilibrium diagram for a  Na2C03-NA2S  system
                                                12,82
     The dissolving tank is a cylindrical  metal  tank  in which green liquor is  .
formed by dissolving the smelt.   The tank  is  vented to provide relief for the
steam produced by the quenching  of the  smelt.
     Green liquor levels are maintained several  feet  below the top of the
tank to allow for expansion and  steam liberation.  High liquor levels result in
carryover of liquor droplets and particulate  matter into the tank vent system.
The tank is agitated with compressed air or mechanical agitation to aid in
mixing.  The water-cooled smelt  spout enters  the tank through a doghouse
enclosure installed on the top of the tank.
     Steam or recirculated green liquor is used  in the smelt spout to shatter
the smelt before it contacts the liquor surface.  Incomplete shattering can
result in violent steam explosions as large droplets  enter the water and
                                     209

-------
 release excessive quantities  of steam.   The  major  causes  of  explosions  in
 smelt tanks  (which constitute a serious  and  potentially dangerous problem)
 are smelt sulfidity,  lack  of  shatter  efficiency, lack of  smelt reduction
 efficiency,  sodium chloride content,  and sodium hydroxide.
      Figure  3-82  and  3-83  show two  designs for smelt dissolving tanks that
                                           O"3
 use water sprays  and/or  steam shatter jets.    Gases that are generated from
 the tank (primarily water  vapor, entrained air, and TRS)  are vented through a
 natural  draft  stack or with an induced draft fan.
      Because of the turbulence in the space  above  the liquor, significant
 amounts  of particulate are emitted  with  the  steam.  The uncontrolled emission
 rate from the  tank varies  greatly depending  on quench rate,  tank design,
 exhaust  volume, and smelt  chemistry.  Some data indicate  a strong relationship
 exists between steam  shatter  flow and carryover.   The location and amount of
                                               84
 steam used also has an effect on TRS  emissions.
      By  proper attention to tank design,  steam shatter jet location, steam
 flow,  and vent control, many  mills  are able  to operate with minimum abatement
 equipment on the  smelt dissolving tank stack.
 3.3.2.2   Sources  of Emission  and Control —
      Particulate  emissions from the smelt tank are controlled by a number of
 low-energy scrubbing  systems.   The  simplest  system consists of a wire mesh
 pad  mist  eliminator equipped  with a back-flush system to  remove collected
 particulate  (Figure 3-84).  The pads  are about 1 ft thick and are generally
made of stainless  steel wire.   Low  vertical  velocities through the pad must
 be maintained to  prevent channeling and/or liquor reentrainment (<20 ft/s).
Water flow is used  on an intermittent basis.   Usage can be determined by time
or pressure drop.    Typical pressure drops are 1 to 2 in.   H^O and back-flush
rates are  3 to 4  gal/1000 acfm of gas.  Expected particulate removal effic-
iencies can range  from 70 to  90 percent.
     Higher removal efficiencies may  be achieved by use of low-energy spray
towers or  packed-bed scrubbers.  The  packed-bed scrubbers are generally used
after low-energy scrubbers to remove TRS emissions.  The scrubbing liquor is
usually weak wash water.
     The use of low-energy entrainment scrubbers with packed beds for par-
ticulate removal  has achieved efficiencies above 95 percent.   Typical  liquor-
to-gas ratios are 4 to 8 gal/1000 acfm at a pressure drop of 4 in.  H20 (Figure
3-85).85
                                     210

-------
                                LEGEND

                    A — dissolving tonk
                    8 — furnace
                    C — vent stack
                    0 — green liquor
                    E — air line for agitation
                    F — dog house
                    G — smelt spout
                    H — circulated green liquor  shatter spray
Figure  3-82.   Smelt  dissolving tank with water sprays.
                                                                   83
                                 211

-------
                                      LEGEND

                          A — dissolving tonk
                          B — furnace
                          C — vent stack
                          D — green liquor
                          E — oir line for agitation
                          F — dog house
                          G — smelt spout
                          H — circulated green liquor spray
                          j — steam shatter spray
                          K — troy
                          L—vent
Figure  3-83.   Smelt  dissolving tank  with  steam shatter jets.
83
                                    212

-------
oo
                       (O
                        c


                        0>


                        CO
                        I
                        00
                        n>
                        in
                       TJ
                        cu
                        D.
                        n>
                        D.
                        (D
                        t/)
                        O
                        _l
                        <
                        _J«
                        3




                        sr
                        3
                        *T

-------
                             Gas Outlet
                                                Eliminator Section
                            Iji^^^iiijII^^^^^^^^^I^J ^M^M
                               Sludge Outlet
Figure  3-85.  Low-energy entrainment scrubber for use on  smelt
         dissolving  tank vent  (Ducon dynamic  scrubber).
                                214

-------
     Low-energy venturi scrubbers also can be used on tank vents.  Common
design pressure drops are 6 to 8 in. HpO at liquor-to-gas ratios of 8 to 10
gal/1000 acfm.  Removal efficiencies can exceed 99 percent.
3.3.2.3  Malfunctions--
     Typical smelt dissolving tank malfunctions involve reduced efficiency of
the abatement system.  Table 3-25 describes malfunctions associated with each
type of control device.
           TABLE 3-25.
  MALFUNCTIONS THAT MAY OCCUR IN SMELT TANK
    PARTICULATE CONTROL SYSTEMS
Control system
         Malfunction
        Result
Mesh pad
Low-energy scrubber
Venturi scrubber
Low water flow
Maldistribution of back
  flush water
Pad pluggage
High gas flow rates

Low water flow
Pluggage of packed bed or
  nozzles
Pluggage or demister
Throat wear
Low water flow rate
Poor water distribution
Pluggage of pad
Pluggage of pad

Channeling and bypass
High pressure drop and
  droplet reentrainment
Low efficiency
High pressure drop and/
  or channeling
Liquor reentrainment
Low pressure drop
Low pressure drop
Low pressure drop
3.3.2.4  Inspection of the Smelt Dissolving Tank Area--
     The smelt dissolving tank area is usually inspected in conjunction with
the recovery boiler area because of the proximity of the smelt dissolving
tank(s) to the recovery boiler and the accessibility to control instrumentation,
which is usually installed in the recovery boiler control room.  The instru-
mentation is normally limited to process-related parameters such as the pro-
duction rate, composition (or density), and temperature of the green liquor,
but some instrumentation may be available for the control equipment.
                                     215

-------
     The  control  equipment used on smelt dissolving tank vents usually consist
of mesh-pad mist  eliminators or low-energy venturi or cyclone scrubbers.
These particulate control devices are well suited to this operation because
a wet saturated gas stream is involved and the particles generated by the
smelt dissolving  tanks tend to be relatively large and easy to remove with
low energy input.  Scrubbing liquor is usually allowed to flow down the stack
walls directly to the smelt dissolving tank.  The use of fiber-reinforced
plastic (FRP) components is common to avoid corrosion problems associated
with high moisture conditions.  In general, the control equipment used on the
smelt dissolving  tank vents is not very complex.
     The vent gases contain water droplets, Na2S, and NaC03 particles generated
by the quenching  of the molten smelt leaving the bottom of the recovery boiler.
The continuous flow of liquid smelt is not, however, allowed to be quenched '
en masse  in the smelt dissolving tank.  Rather, steam shattering nozzles are
typically applied to the smelt spout to form particles of smelt that are
easier to cool and dissolve because they have a larger surface area.  As the
molten smelt is cooled and steam evolves under relatively turbulent con-
ditions, however, smaller particles of smelt are carried out of the dissolving
tank in the vent  gases.  These are the particles that the control equipment
is attempting to  collect.
     The three levels of inspection that have been defined for this area are
outlined below.   Limitations of accessibility to measurement points will
often limit the inspector to a Level  II Inspection with some of the elements
of a Level III Inspection.   Much depends on the availability of instrumen-
tation and access to the control  equipment.
     Level I Inspection—Observing visible emissions is of limited value for
the smelt dissolving tank stacks.   The three general limitations are 1)
uncombined water vapor produced by quenching of the smelt usually causes the
formation of a condensed steam plume,  which forces observations to be made
beyond the point of steam plume dissipation; 2) the stack outlet is usually
near other outlets associated with the recovery boiler operation, which may
interfere with the ability to read the opacity'of the individual plume; and
3) it is difficult to select an observation site that will  provide optimal
background and contrast conditions and establish the proper sun-source orien-
tation without interference from other sources.
                                     216

-------
     Observing visible emissions observation is useful,  however,  when large
changes in the opacity occur.   Such changes may indicate a change in the
operation of the smelt dissolving tank or control  equipment.   For example,  a
change in the positioning or pressure of the shattering  nozzles may cause
a shift in the size of the smelt particles formed.  If smaller particles are
formed, they may be more difficult to collect in the control  equipment at a
fixed energy input level and greater penetration may occur.  In addition, a
shift in particle size may enhance the light-scattering characteristics of
the particulate so that opacity would increase as a result of both finer
particulate and higher concentrations.  A failure in the control equipment
might produce similar results.  Thus, the visible emissions observation of
the Level I Inspection may be used as a screening tool to determine areas
that may warrant further investigation.
     Level  II Inspection—The Level  II Inspection involves a combination of
observation of  the opacity of the stack plume  and a physical check  of  the  air
pollution control equipment to  confirm operation  (although equipment operating
parameters  are  not measured).   Operating  data  from available instrumentation
are  obtained  for both  the process and the control equipment.   Typical  process-
related information  that  should be recorded include green  liquor production
rate (tons/hour, gallons/minute, or  gallons/hour),  liquor  composition  or
density,  liquor temperature,  and smelt dissolving rate  (from  recovery  boiler).
Typical control equipment parameters that may  be  monitored include pressure
 drop,  stack temperature,  and  the flow rate, pressure, and (in  some cases)
 temperature of the  scrubber water.   Because no measurements  are made of these
 parameters  and the  inspection includes  only visual  confirmation of component
 operation (pumps,  fans, etc.),  the inspector must base  his or her subjective
 assessment of the  accuracy  of these data on this  monitoring effort.  If the
 instruments providing the data are inadequate or malfunctioning, the result
 may be an incorrect assessment of the control  equipment compliance status.
      To the extent possible,  the values obtained during the inspection should
 be compared with values from the design, baseline,  or previous  inspections to
 determine if there has been a significant change in performance or in the
 number of equipment malfunctions.   With the low-energy equipment that is
 being  used, performance can change significantly with only a  slight shift in
                                      217

-------
 pressure drop.  For example, a change from a 6-in. to a 4-in. H-O pressure
 drop in a low-energy venturi is significant, whereas with much higher operating
 pressures this shift might not be considered significant.  If particle size
 is not very small, however, a shift in pressure drop may not significantly
 affect emissions.  It is usually the identification of the causal factors of
 the parameter change that determines the magnitude of performance changes.
 These causal factors are not always easily identified through a Level II
 Inspection.
      Level  III Inspection—The detailed inspection procedures associated with
 a Level III  Inspection are slightly different for each control  device used to
 minimize the emissions from the smelt dissolving tank, but all  of the control
 devices have common elements.   The characteristics of each type of control
 equipment are discussed to aid the inspector in the diagnosis of operation
 and/or maintenance problems.
      The operating parameters  that are common to all  the pieces of control
 equipment include smelt production rate,  green  liquor production rate,  liquor
 composition  or density,  and*liquor temperature.   The  control  equipment  param-
 eters the inspector should obtain  include the scrubber pressure drop, stack
 temperature,  scrubber  water flow rate,  scrubber water pressure,  and,  in  some
 cases,  scrubber water  temperature.   These parameters,  along with the  obser-
 vation  of visible  emissions, are the same components  that makeup a Level  II
 Inspection.   The  difference between  Level  II  and III  Inspections is that in
 the latter the inspector actually  measures some of these parameters.  In
 addition, the  inspection is  keyed  toward  identifying  operating  problems  that
 characterize  control equipment performance.
     A most useful  parameter for diagnosis of the  control  equipment operation
 is the measurement  of volumetric flow rate of the  gas  into the  control equip-
ment.  The most accurate method of obtaining  this  parameter is  to  select a
suitable location for performing a Pi tot  traverse  and  determining  volumetric
flow rate.  The gas density will be  somewhat  less  than air density because
the gas will  contain a substantial fraction of water vapor at (typically)
saturated conditions.  The volumetric flow rate will be  related  to the quantity
of air inle'akage around  the smelt dissolving  tank, the production rate of
smelt, and the smelt temperature leaving the  furnace  (affecting the steam
evolution rate).  Using fan operating parameters to determine gas volume

                                     218

-------
(where fans are used) is usually impractical because the density of air and
water vapor differs significantly and because entrained water droplets can
increase the energy requirements of the fan.
     The value obtained by use of the Pi tot tube should be compared with the
design gas volume or a value obtained from a previous stack test.  If pro-
duction rates are comparable, the volumetric flow rates should be similar
(within +15 percent) unless changes have occurred elsewhere in the system.
     Another key parameter that should be measured is pressure drop.  Although
this parameter is typically monitored by a magnehelic gauge, the taps should
be removed from the sampling locations and cleaned, and the inspector should
check the measurements with his or her own gauge.  In some instances, the
pressure taps are connected to recording and controller instrumentation that
regulate water flow rates and alarm systems.  Taking care not to interfere
with these systems, the inspector should measure pressure drop through sepa-
rate taps that do not interfere with active systems.  Before removing taps,
the inspector should check with plant supervisory personnel concerning any~
union rules that might require union personnel to perform this task.
     The principal operating problem to be detected in the mesh-pad control
and low-energy venturi scrubbers is one of improper distribution of scrubbing
water or the gas stream within the control device.  These maldistribution
problems tend to allow a portion of the gas stream to bypass the zone where
particulate collection mechanisms are operating.  In addition, maldistribu-
tion problems are often more crucial in the operation of this equipment
because such small energy inputs are typical.  For,example, mesh-pad mist
eliminators tend to operate at around a pressure drop of 2 inches ^0, venturi
scrubbers operate with a pressure drop of 6 to 8 inches H20, and cyclone
wet-fan scrubbers operate with a pressure drop of between 4 and 8 inches  H20.
These small energy inputs make the control equipment sensitive to maldistri-
bution problems, whereas larger pressure drops tend to reduce the sensitivity
to distribution problems.  If the scrubbers are operating properly, however,
larger energy inputs are usually not necessary.
     A change in pressure drop in these low-energy devices is normally signif-
icant and indicates that the performance of the control equipment has changed.
Fortunately, the particulate size is large  by air pollution control standards,
which helps reduce the sensitivity to changes in pressure drop.  The  purpose

                                     219

-------
 of the Level III Inspection is to assess the causes of these changes and
 determine their significance with regard to emission rates.
      For mesh-pad eliminators, the usual indication of operating problems is
 an increase in pressure drop, usually caused by an increase in the pad's
 resistance.  Most mesh-pad systems use continuous back-flushing sprays through -
 the bed to minimize buildup of particulate, but some systems use a pressure
 drop or timer-controlled back-flush system.  The presence of the mesh pad and
 the water flowing through it normally place some resistance on gas flow.  If
 the back-flush system fails or water distribution is not adequate, particulate
 buildup will occur in the bed and cause the pressure drop to increase.
      Failure of the water spray system will usually cause a relatively rapid
 increase in pressure drop.   A decrease in water flow rate to near zero flow
 (because of line or nozzle pluggage, increased water pressure, or an inoperative
 pump) also could be indicative of mesh pad pluggage resulting from failure of
,the water spray system.   Insufficient water flushing volume or inadequate
 coverage of the mesh pad may cause a partial pluggage to occur over a longer
 time period.  The existence of localized pluggage will  cause an increase in
 the local gas velocity through the bed, which will increase the pressure drop
 and could increase the penetration and reentrairiment of droplets and particles
 through the control  equipment.   When conditions permit and plant personnel
 allow,  the access door to the mesh pad should be opened to determine if the
 spray nozzles are providing adequate coverage.
      Low-energy venturi  scrubbers also are very sensitive to improper water
 distribution in the throat area.   In a practical sense, these low-energy
 scrubbers,  which operate in the range of 6 to 10 inches pressure drop, are
 more sensitive to slight changes  in pressure drop and water distribution than
 venturi  scrubbers operating in  the 18- to 45-inch range.   The low energy
 associated with these scrubbers means the available impaction energy is low,
 and the  poor distribution allows  the impaction  zone to  be bypassed.   These
 scrubbers typically  use  spray nozzles to achieve the gas-liquid contact
 required in the scrubber throat,  as the low-energy impact may not be high
 enough to achieve gas atomization of the scrubbing liquid and good throat
 coverage.  -The relatively large particle size from this emission source,
 however,  favors low  energy  input  to the scrubber.
                                     220

-------
     When inspecting low-energy venturi scrubbers, the inspector should check
the gas flow rate, the scrubbing liquid flow rate, and the pressure drop.   A
small change in pressure drop (i.e., a change of 1 inch H20 or more) should
be considered significant, particularly if it is a decrease.  It is difficult
to discuss all of the possible malfunctions and their effect on performance.
An increase in the pressure drop, liquid-to-gas ratio, or throat velocity,
however, will generally improve performance, provided good liquid-to-gas
contact is established in the scrubber.  There is no easy method by which an
inspector can establish whether good water distribution is being provided to
the scrubber during normal operation unless separate valves are provided for
each nozzle.  If the scrubber is equipped with individual valving to allow
for individual nozzle inspection and cleanout, each nozzle could be turned
off, removed, inspected, cleaned (or replaced, if necessary) and returned to
service.  Turning off a nozzle will produce only a slight change in water
flow rate and water pressure, but a change in pressure drop should be evident.
If any nozzle is turned off without having any effect on the scrubber per-
formance, that nozzle may be plugged or severely eroded and could be causing
a maldistribution of the scrubbing liquor.
     Many plants use a cyclone spray scrubber.  This scrubber, actually a
hybrid of .a cyclone spray scrubber and a wet fan collecting device, is also a
relatively  low-energy device with typical pressure drops of 6 to 12 inches.
The device  takes advantage of particle inertia to promote impaction of the
particle with the walls of the scrubber and the fan rather  than using  impac-
tion of the particle onto water droplets in the gas stream  (although this
mechanism does play a role in the collection of the particulate).  The water
applied in  this scrubber  is applied primarily as a particulate-transporting
medium once the particulate has reached a surface inside the  scrubber.  This
scrubber  is much less susceptible to pluggage or problems with improper water
distribution  than the scrubbers previously  discussed.
     The  gas  is usually introduced  tangentially into  the scrubber  to remove
the  large particles and liquid droplets.  A spin vane  system  also  may  be  used
at the  inlet  to begin the  inertia!  separation process.  Water is introduced
into the  scrubber to minimize pluggage and  erosion of the spin vane as well
as to provide a transport  medium for the  large  particles collected in  this
                                     221

-------
 portion of the scrubber.   The gas stream is then passed to a wetted radial-
 blade centrifugal  fan,  where all  but the smallest particles are collected.
 Significant inertia!  forces are applied to the particles and water droplets,
 and their inertia  tends to force them to the outer wall  of the fan housing,
 where the liquid is drained from the scrubber.   Some particles become impacted
 on the fan blades  and are removed by the water applied to the fan  wheel,  and
 some small  droplets are created by the shattering force applied by the rotating
 fan wheel.   The remaining water droplets in the gas stream are then separated
 by cyclonic action through a spin vane separator.
      Most of the energy applied to the particles is supplied by the rotation
 of the fan.   The fan  must provide the energy to overcome the energy losses  of
 the ductwork,  the  cyclonic separation of the particulate and water droplets,
 and the acceleration  of the water and particulate in the fan.   The usual
 procedure is to determine the pressure drop across  the entire scrubber from
 inlet to outlet.   Using fan curves to calculate gas volumes  through the
 scrubber is  not possible  because  of the direct  application  of water to the
 fan blades.   The use  of Pitot traverses is  usually  not possible either because
 of the presence of cyclonic flow  after the  scrubber.   Thus,  the only  indicators
 available to the inspector for assessment of performance are pressure  drop,
 water flow  rate to the  scrubber,  and fan motor  current.
      Although  less susceptible to pluggage  than mesh-pad scrubbers,  pluggage
 that occurs  in  cyclone  scrubbers  will  generally increase the pressure  drop
 across  the  scrubber because of an unbalancing and increased  resistance to gas
 flow through the scrubber.   A decrease in liquid  to the  fan  may not have much
 effect  on the pressure  drop across  the scrubber,  but such a  decrease  tends  to
 be  indicated  by a  decrease  in  the fan  motor  horsepower because  of  the  apparent
 decreased "gas"  density on  the  fan.   Thus, while operating parameters  are
 limited,  those  that are available may  indicate  scrubber  performance changes.
     As mentioned  previously,  the  low energy input  of  the scrubber  limits the
 ability of the  scrubber to  collect  fine particulate; however, the  particulate
 generated is generally large.   Changes  to smaller particle size  because of
changes in smelt characteristics  or  positioning,of  steam smelt  shattering
nozzles may" cause  problems  in particulate collection efficiency, and when
combined with other malfunctions, can  cause performance  to deteriorate sig-
nificantly.
                                    222

-------
3.4  CAUSTICIZIN6 DEPARTMENT
     The two major emission sources in the causticizing department are the
slakers and the lime kilns.  The major source of particulate emissions in the
causticizing department is the lime kiln which is typically controlled by a
venturi scrubber.  The malfunctions associated with lime kilns can generally
be divided into those that are process-related and those that are related to
scrubber performance and operation.  Section 3.4.4 provides a detailed dis-
cussion of lime kiln and venturi scrubber malfunctions.  Section 3.4.5 pro-
vides a discussion of the inspection procedures for lime kilns and slakers.
In general, Level III inspections are usually limited by production schedules,
availability of instrumentation, and accessability to process and control
equipment.
3.4.1  Process Description
     The economical operation of the alkaline pulping process requires the
recovery and reuse of the sodium and sulfur compounds used in the c'ooking
process.  The liquor produced by the smelt dissolving tanks  (green liquor)
contains a mixture of sodium carbonate and sodium sulfide.   Because sodium
carbonate cannot be used in the cooking process, it must be  converted to a
usable form, NaOH.
     The causticizing process involves the conversion of Na2C03 to NaOH
through the use of calcium oxide  (CaO)  (Figure 3-86).2  Causticizing occurs
in two steps:  1) reaction of the  calcium oxide with water and 2) reaction of
the  calcium hydroxide with the  sodium carbonate.
     The first reaction', which  is  referred to as slaking,  liberates a con-
siderable amount of heat.  The  reaction  is written as  follows:
        '  CaO +  H20 ->• Ca(OH)2 + 486  Btu/lb
The  second  reaction, which is referred  to as  causticizing, yields sodium
hydroxide and calcium carbonate.   This  reaction  is written  as follows:
          Ca(OH)2 + Na2C03 *  2NaOH + CaC03
 In normal practice  both these reactions  occur simultaneously in  the  slaker
 tank.   Because the  second  reaction is  reversible,  all  of  the sodium  carbonate
 cannot be converted  to  sodium hydroxide.  The completeness of the reaction is
 defined as  the causticizing  efficiency.
                                      223

-------
GREEN LIOOOR
 TTJ^
                                                             o OC
            Figure 3-86.  Typical  causticizing flow diagram.
                                      224

-------
     The  efficiency may  be  determined  by  the  following equation:
           Efficiency  (%) =
                               NaOH
NaOH
                                   Na2C03
              x 100
The efficiency  is a function of  the concentration of sodium sulfide in the
green liquor.   Higher concentrations of sodium sulfide reduce conversion
efficiency.  Unreacted compounds represent an unreacted circulating load in
the liquor system.  Sodium sulfide in the liquor hydrolyzes to form sodium
                                 85         *
sulfide and sodium hydrosulfide.
     For recovery of the calcium used in the slaking process, calcium carbonate
is separated from the"white liquor, dewatered, and disassociated under intense
heat.  The reaction is referred  to as lime burning or calcining and proceeds
according to the following reaction:
          CaCOo •* CaO + COp

     The causticizing process is composed of many subprocesses, including
green liquor preparation, white liquor preparation, and lime mud handling.
3.4.1.1  Green Liquor Preparation--
     Green liquor produced by the smelt tank contains insoluble impurities
known as dregs, which must be removed.  Dregs, which are composed of carbonaceous
                                        2
material, silica, and metallic sulfides,  are removed in the green liquor
clarifier through settling and decanting.  The solution is washed in the
dregs washer, where the dregs are removed.   The wash water is then pumped to
               oc
the mud washer.    .
3.4.1.2  White Liquor Preparation--
     Clarified green liquor is pumped to the green liquor storage tank for
introduction into the slaker.   Calcium oxide is introduced into the slaker
(reburn plus pebble lime) and reacts to form sodium hydroxide and calcium
carbonate according to the previously stated reactions.  The slaker contains
a clarifying section in which unreacted material  (grit) is removed by a
                              87
mechanical  rake (Figure 3-87).
     Lime feed to the slaker is generally no more than 85 percent CaO.  The
reaction is carried to equilibrium in the causticizers (up to three may be
used).   White liquor containing lime mud is pumped from the causticizers to
                                     225

-------
VtNT STACK OPCNIMC"
                                  .FLANGED HPt CONNECTION FOR
                                  OVCRfLOW CONNECTION TO
                                  CITHCR3IDC OF MACHINE
OORRCOMIXCR
 RING SEAL-
awffis"
Bath*
                                                                                   .CONNCCTMC HOC
                                                                                    FMNT KAKC MANCCR
                                                                                     FRONT LINK
                                                                                       •GKIT OIKHAKCC
                              -

                             "- -LI
                                         SECTIONAL CievATION
                Figure  3-87.   Slaker-classifier  used  in  typical
                                  causticizing  pi ant.87
                                             226

-------
the white liquor clarifier, where the mud and liquor are separated.  The
clarified liquor is then pumped to the digesters.
3.4.1.3  Lime Mud Washing--
     Because the lime mud slurry contains sodium, which can adversely affect
the calcination process, it is washed in dilution washers to reduce the
sodium and sulfide content.  Waste wash liquor is returned to the smelt
dissolving tank as weak liquor.  Lime mud washers may be decanting, belt-
filter, precoat-filter, or pressure-filter washers.  The lime mud must be
dewatered before being introduced to the lime kiln.  Centrifuge or vacuum
filter dewatering systems are typical.
     Final soda content is between 0.1 and 2.5 percent by weight (as Na?0),
                                                      88
and dewatered mud is between 60 and 70 percent solids.
3.4.1.4  Calcining—
     Calciriing takes place on fluid-bed calciners or in rotary, direct-
fired kilns.  Rotary kilns used in kraft mills are 8 to 13 feet in diameter
and 100 to 400 feet in length.  The kilns slope downward from the feed end to
the discharge end at an angle of about 3/4 in. per foot of length.  The feed
mud is moved through the kiln by slow rotation, usually between 0.60 and 1.33
rpm.  The rate of material movement down the kiln is determined by the angle
of the lime mud respose, kiln length, kiln slope, rotation speed, and kiln
diameter.
     Processes carried out in the kiln are water evaporation, mud heating,
and calcination.  Feed end temperatures are between 300° and 500°F and dis-
charge end temperatures are between 2000° and 2400°F.
     Energy for calcining is supplied by direct firing countercurrent to the
material flow.  Fuels are normally residual and distillate oils, natural gas,
and coal, but waste oils, turpentine, and tall oil soaps also may be used.
     Because of the presence of inert material in the lime mud, product yield
(CaO) will not equal theoretical values.  A typical mud may contain up to 15
percent inert material.  Normal yield, based on inlet feed weights, may be on
the order of 45 to 50 percent.  In well-operated systems, fuel rates are in
the range of 7 to 7.5 x 106 Btu/ton of product.
                                      227

-------
3.4.2  Emission Sources
     Particulate emissions from the kiln are the result of particle suspen-
sion and entrainment due to the flow of combustion gases and carbon dioxide.
Fine particulate is generated by the vaporization and condensation of sodium
based compounds.  The amount of fine particulate generated is related to the
soda content of the feed.
     Secondary and primary air for fuel combustion are provided at the burner
end of the kiln.  The burner-end hood is held at a slight negative pressure
(0.1 in. HpO).  The air supply rate is controlled to maintain an exit gas
oxygen content of 1.5 to 3.0 percent (7 to 15 percent excess air).  Feed-
end drafts can be significantly higher, depending on kiln diameter and length
and the weight of chains used.
     Table 3-26 presents a mass balance of a typical 260 tons/day kiln.  The
dust in the mud feed is recycled lime dust from primary dust collectors at
                                                                    ~i
the kiln exit.
                TABLE 3-26.  TYPICAL LIME KILN MASS BALANCE89
Lime mud feed, tons/day
     CaC03
     Dust
     Inerts
     Water
Product rate, tons/day
     CaO
     Inerts
Emission rate, tons/day
     CaC03
     CaO
Heat input, 106 Btu
Heat input/ton product, 10  Btu
Kiln size, ft
               3
Kiln volume, ft
Kiln gas volume, acfm
Kiln gas temperature, °F
Retention time, min
 417.857
  28.331
  26.000
 101.366

 234.000
  26.000

  28.331
   0.000
  85.069
   7.8525
12 x 275
  24,971
  68,000
    450°
 129.430
                                      228

-------
     Lime mud crystals that precipitate from the white liquor in the causti-
cizers have Na2S that cannot be removed during mud washing.  The removal rate
depends on the sulfidity of the wash water and the ratio of soluble to in-
soluble sulfide in the mud.  As a result, Na2$ is carried into the kiln, where
it is exposed to an atmosphere containing approximately 18 percent C02.   The
C02, adsorbed on the mud surface, lowers the mud pH, which allows the sulfide
to be converted to H2S.  The rate of H2$ that is released is a function of the
Na2S content of the mud, mud moisture, 02 content of the gas stream, and the
flue gas temperature.  Figures 3-88, 3-89, and 3-90 show the effect of these
variables on H2$ emission levels.
3.4.3  Control
                                 88
     The control of particulate emissions from the lime kiln is provided
almost exclusively by venturi scrubbers.  Although impingement plate scrubbers
and ESP's have been used in some locations, the venturi scrubber offers
significant economic advantages.  First, venturi scrubbers are more compact
in size than other scrubbers of comparable efficiency, which means less
capital cost and space requirements.  Second, plugging is less of a problem
than it is with other scrubbers.  Although plugging and erosion of lines and
nozzles are possible in a venturi scrubber, they can tolerate much higher
levels of solids in the scrubbing liquor (10 to 20%) than that which can be  •
tolerated by an impingement plate scrubber (1 to 3%).  This usually means
less clarifying, treatment, and makeup water requirements for the venturi
scrubber.  The single largest disadvantage of the venturi scrubber is the
energy requirements to maintain a high, pressure drop.  Pressure drops between .
20 to 35 inches hUO, which require large fans and motors, are not uncommon
for venturi scrubbers applied to lime kilns.
     Most venturi scrubbers are constructed out of 316 and 316L stainless
steel to prevent corrosion.  Most designs also have a "flooded elbow" variable
throat that allows for additional pressure drop control.  The flooded elbow
at the exit of the venturi scrubber divergent section provides a trough of
water to capture the larger water droplets and particulate that exit the
scrubber and prevents abrasion of the elbow at the turn.  The throat of the
venturi may be circular, rectangular, or oval, and the throat area can be
                                      229

-------
                  f  "
                  o
                  SI
                        MOISTURE - 0.4 g/i •( *y CaCO.
                        TEMPERATURE- SOO*F
                        OXYGEN- 2,3X
                                                       TOO  »OO
                                   .S IN LIME MUD
                                   ?b of dry CoCO,
Figure  3-88.
H2$  emission from the  lime kiln related to  Na?S  level
             in the lime  mud.88                   '
                   en
                 #
                                        •M.S -1001*4 ,< +,
                                        MOSTU*E-0.4e/l| of »j OCO,
                                        Tf MR.- SOO V
                              OXYGEN IN FLUE GAS — %
Figure 3-89.   H2S  emission related  to percent 02  in the  flue gas.
                                                           88
                                     230

-------
                        100-

                        90-

                        80-

                        70 -

                        60

                        so!
                                            N«,S-ioom/o of ft c«cc,
                                            TEMP. - 800 V
                                            orroEN- 1.6%
                                 ta       a*
                                  MOISTURE • LIME MUD
                                  Orom/grom of *y CoCO,
Figure 3-90.   H2$  emission related to the moisture content in the lime mud.
                                                                            88
controlled by  hinged  plates  in the throat or by a "bob" that moves up and
down in the effective throat area.   Pressure drop is usually controlled by
controlling the  throat area  (hence,  throat velocity), not by controlling the
water flow rate  to  the scrubber.
     Water is  usually introduced  to  the scrubber through an overflow Weir
system above the throat to avoid  the use of nozzles, which may become plugged
or eroded as a result of the solids  in the scrubbing water.  Throat coverage
by the water is  somewhat low because the water flows down the walls of the
scrubber to the  throat.   This problem is more than offset, however, by the
avoidance of nozzle problems.   Some  designs do use nozzles to introduce water
droplets into  the throat, but adequate maintenance capabilities must be
provided for cleanout and inspection of the nozzles while the scrubber is
operational.
     Particle  impaction on the water droplet is the dominant collection mech-
anism in the venturi  scrubber.  The  high-velocity gas entering the throat
(gas atomization) shatters the water introduced into the scrubber into "fine"
droplets with a  large effective surface area.   These water droplets provide
impact targets for  the particles  that cannot follow the gas streamlines
around the droplets,  and the particles collide with the water droplets.
                                      231

-------
Particles with diameters of 25 to 75 ym are much easier to collect than
particles in the 5- to 10-ym range.  Compared with the gas stream, the water
droplets have essentially no velocity when introduced into the scrubber
throat.  Impaction is the major collection mechanism at the point of intro-
duction into the scrubber throat.  The velocity of water droplets, however,
accelerates through the throat of the scrubber, and the effectiveness of the
impaction mechanism decreases as the water velocity approaches the gas
velocity.  Smaller water droplets, higher throat velocities, higher liquid-
to-gas ratios, and the resulting higher pressure drops usually mean higher
collection efficiencies.
     Diffusion is another collection mechanism in the venturi scrubber.  This
mechanism applies only to very fine particles, generally less than 0.05 ym,
where Brownian motion influences the particle motion and particles tend not
to follow gas streamlines.  This mechanism is usually operative in the
divergent section of the venturi where water droplets in the gas stream are
travelling at nearly the same velocity.                               •
     Unfortunately, neither impaction nor diffusion is very effective in a
certain range of particle sizes  (0.1 to 1 ym).  In this range, the particles
are small enough to follow gas streamlines around the droplets, but are too
large to allow diffusion'to be an effective collection mechanism.  If a
significant portion of the particles entering the scrubber falls in this
range, very high pressure drops may be required for adequate control of
particulate emissions.  Particles in this size range are the ones that most
effectively scatter light and cause an observable opacity.  Lime mud with a
high soda content  (as a result of inefficient washing) can produce a large
quantity of particles in this size range.  Increased pressure drop or process
modification may be the only solution to this problem.
     In many mills, emissions from the lime slaker are often  ignored and no
controls are provided.  The emissions of lime dust and steam  from the lime
sleker and the causticizing of green liquor to white liquor can be controlled
with low-energy scrubbers.  When  slaker emissions are controlled, the usual
equipment consists of cyclone scrubbers or dynamic scrubbers  using cyclone
scrubbers with a wet fan arrangement similar  to that used  for smelt  dissolving'
tank vents.
                                      232

-------
     Scrubbers in this application usually need only 4 to 6 in. of pressure
drop to be effective. .The dominant collection mechanism is inertial separa-
tion; i.e., inertial forces on the particle cause it to separate from the gas
stream, either in the cyclone scrubber or in the wet fan arrangement.  Scrub-
bing liquid may be water, weak white liquor, or clarified green liquor.  The
scrubbing liquor is usually returned directly to the lime slaker for use in
the lime recycle loop.
     Scrubbers used on causticizers and lime slakers usually must be hooded or
placed in covered enclosures to ensure adequate capture efficiencies and to
minimize ventilation requirements.  Scrubber components are normally stain-.
less steel to minimize corrosion problems.
3.4.4  Malfunctions
     The malfunctions and problems in the lime kiln area can generally be
divided into those that are process-related (i.e., kiln operation) and those
that are related to scrubber performance and operation.  The process-related
problems generally will  influence the 'scrubber performance by affecting both
the temperature and the particle size entering the scrubber.  These problems
may be the result of equipment failures, but they are most often caused by a
process change.
     The most common problem in the lime kiln process is inadequate washing
of the lime mud before its introduction into the kiln.  As discussed in Sec-
tion 3.4.1, an interaction occurs as a result of material production rates,
design parameters, yield efficiencies, and quantities of inerts in the
caustizing area.  From an air pollution control standpoint, problems begin
when lime kiln mud from the causticizing of green liquor to white liquor is
not washed effectively before it is introduced into the kiln.  Residual
sodium compounds, particularly sodium sulfide (Na2S) and sodium hydroxide,
remain in the lime mud after causticizing, and the mud must be washed and
filtered to remove as many of these constituents as possible.  Because it is
impossible to remove all of these compounds, the lime kiln can become a
significant generator of fine particles and H2S if process conditions are
right.
     The sodium compounds are easily volatilized in the high-temperature
flame end of the kiln, and they tend to leave the kiln as an uncondensed

                                     233

-------
alkali material.  Unless the temperature of the gas stream is reduced before
it reaches the inlet of the scrubber, most of the alkali will remain in the
gas phase until it is quenched in the scrubber throat.  Unfortunately, alkalies
tend to form in the 0.1- to 1-ym range, which makes them difficult to collect
in a venturi scrubber.  One possible solution is to quench the gas stream by
use of a presaturator.  This allows the particles to grow before they enter
the inlet of the scrubber.  A-better solution to this problem, however, may
be to change the washing characteristics to reduce the sodium carryover in
the lime mud.
     The generation of fine particulate is a symptom of sodium-based compound
carryover, but the principal compound in the lime mud that is related to
potential H2S emissions is Na2S.  (See Figure 3-88.)  The conditions of high
C02 levels and high temperatures can promote the breakdown of Na2S to form
H2S.88  The concentration of Na2S in the lime mud is not the only variable
that affects the generation of H2$, however.  The presence of oxidizing
conditions  (02) and the temperature also affect H2S emission rates.  Low
temperatures or low 02 contents enhance the formation of H2S, whereas high   v
temperatures and higher 02 levels cause equilibrium shifts to form S02.
Thus, short kilns with higher operating temperatures have less potential to
emit H2S than do longer kilns at the same production rate and lower  tempera-
tures.  Kilns operating with oxygen enrichment to increase production rates
may also produce less H2S because of high .operating temperatures  in  the kiln.
If the 02 at the exit end of the kiln  is 0. to 5 percent, sufficient  excess
02 must be  provided to avoid H2$ generation if residual oil  is burned.
      In addition to inefficient or  insufficient lime mud washing, potential
contamination can come from the washing liquor  itself.  If the incoming
washing liquor  is contaminated with sodium  compounds, particularly  Na2$ from
other portions  of the mill, then washing will be  unable to reduce the  content
below the level of  the  incoming wash solution.  The eventual  solution  to  this
problem may require repiping  the supply lines to  avoid  the contaminating
source(s).
      Scrubber  operating  problems typically involve  failure to maintain  pres-
sure  drop,  difficulty in  supplying  scrubber liquor  to the  scrubber,  or a
buildup  of  solids  in  the  liquor.   Because  some  of these problems  are inter-
related,  identifying  a  specific problem can be  difficult.
                                      234

-------
     Most scrubber controllers use pressure drop as an indicator of perform-
ance.  Failure to maintain the scrubber pressure drop at a preset point can
cause operating problems at either extreme.  High pressure drops can cause a
fan limiting condition that prohibits the fan from moving the required quantity
of gas.  This results in fugitive emissions from the kiln, which operates
under a slight negative pressure.  Other failures result from the inability
to maintain adequate pressure drop through the scrubber.  This condition may
be caused by failure of the pressure drop controller system, a water distri-
bution problem, or the abrasion and erosion of scrubber internal components,
all of which usually decrease scrubber performance.  The latter two problems
can usually only be identified by an internal inspection, whereas a failure
in the controller system can normally be identified by a lack of response to
a change in pressure drop set point.
     The easiest method for controlling pressure drop in a variable-throat
venturi scrubber is by controlling the throat area and the effective throat
velocity.  Although control of the water flow rate is an alternative means
of controlling pressure drop (usually the only alternative in fixed-throat
venturi scrubbers), it is not often used because it is more difficult and
less sensitive to error adjustment.  Controlling both the throat velocity and
the water flow rate is generally not successful because of the interactive
relationship of the two parameters on pressure drop.  Attempting to control
both parameters typically results in unacceptable or unstable oscillations in
pressure drop or extremely slow response times.
     Water distribution is usually not a problem in these scrubbers because
nozzles are seldom used.  When nozzles are used, however, provisions must be
made to isolate and extract each nozzle for inspection, cleaning, and replace-
ment during normal operation.  Although inspection of each nozzle may cause a
temporary maldistribution of liquid in the scrubber, it is better than a
longterm failure.  The maldistribution of liquid over a period of time can
cause excessive abrasion of the throat components resulting from localized
increases in velocity with abrasive particulate.  Inadequate covered throat
area will allow the gas to "channel" to the area of least resistance.
     The solids levels in the recycled scrubber liquid can cause erosion of
pump impellers and pipes.  Wear of pump impellers can decrease the volume and
velocity of scrubbing liquid to the scrubber, which can decrease scrubber
                                     235

-------
performance and cause pluggage of liquid supply lines.   High solids levels
can also cause nozzle pluggage in addition to nozzle erosion.
     High solids levels also can cause problems with resuspension of particu-
late in the gas stream where the hot gas stream contacts the liquid at the
throat.  The evaporation of the liquid allows both the suspended and dissolved
solids to be resuspended in the gas stream.  The suspended solids level  can
be controlled by using a bleed-off stream with fresh water makeup to prevent
high concentration levels.  The problem also can be helped somewhat by quench-
ing the gas stream with fresh water.  Use of water with high dissolved solids
content can allow the formation of a submicron fume that is difficult to
collect at normal pressure drops.  A bleed-off stream and fresh water makeup
can also help to reduce problems with high suspended solids.  A material
balance typically shows that a site specific equilibrium level can be established
at which particulate will begin to become resuspended in the gas stream.  It
is desirable to maintain solids levels below those levels that increase resus-
pension.
     Malfunctions in the lime slaking area are generally limited to inadequate
distribution of the scrubbing liquid in the scrubber.  Because these are
low-energy devices, the malfunctions tend to be related to scrubber pump
failure or nozzle pluggage.  The other failure mechanism is generally related
to the inability to capture the emissions because hatches, covers, or hoods
are not properly aligned.
3.4.5  Inspection Procedures
     Control of the lime kiln particulate emissions is usually accomplished
through the use of a venturi scrubber.  Although other types of scrubbers are
sometimes used, most of the newer lime kilns have venturi scrubbers operating
in the range of 20 to 35 inches pressure drop.  For this reason, the discussion
of the inspection techniques in the lime kiln area is limited to venturi
scrubbers.
     As discussed previously, the lime kiln provides the second chemical
recycle loop in the kraft pulping process, which allows green liquor to be
converted to white liquor for use in digestion of wood chips.  The particulate
generated by the lime kiln is primarily calcium carbonate and calcium oxide.
Energy requirements for the scrubber depend on the size distribution of the

                                    236

-------
particles leaving the kiln and entering the scrubber.  Effective collection
of smaller particles requires higher pressure drops.  Scrubbers are usually
fabricated of 316 stainless steel and are designed with a flooded elbow to
minimize wear at the exit of the scrubber.
     The three levels of inspection for this control device are defined.
Ability to conduct a complete Level III Inspection is usually limited by
production schedules, availability of instrumentation, and accessability to
equipment.
     Level I Inspection—The Level I Inspection of the lime kiln area has the
same limitations as those for other wet plumes:  1) uncombined water vapor
caused by evaporation in the scrubber usually causes the formation of a con-
densed steam plume, forcing observations to be conducted beyond the point of
steam plume dissipation; 2) the selection of an observation point may not
provide good or optimal background and contrast conditions for the establish-
ment of the proper sun-source orientation; and 3) when several kilns and
scrubbers are located close together, the plumes may interfere with each
other.  In some cases, the exhaust from several scrubbers may be combined in
a single stack.
     All scrubbers will not exhibit an attached plume at the stack exit, even
though the gas stream is at saturated conditions.  The energy input by the
fan in high-energy scrubbers may provide just enough energy to revaporize the
condensed water droplets and cause an unattached steam plume to form that may
facilitate opacity readings at the stack exit.
     A change in opacity from the baseline condition or a previously established
operating condition indicates that some parameter or group of operating
parameters has changed, which warrants a more detailed inspection of the
equipment.  Therefore, a change in opacity levels may be used to establish
the need for conducting more detailed inspections.  Without further operating
data, however, the cause of the opacity change is difficult to assess with a
Level I Inspection.
     Level II Inspections—The Level II Inspection entails a combination of
visible emissions observation and a physical check of the air pollution
control equipment by use of plant instrumentation to confirm its operation.
Usually the inspector does not take measurements in a Level II Inspection.

                                     237

-------
If instrumentation is properly located and operated, however, the data pro-
vided can be useful in evaluating the performance of-the venturi scrubber.
Operating data from both the process equiment and the control equipment
should be obtained during the inspection.  Typical data that should be col-
lected include kiln production rate (as CaO), lime mud feed rate, fuel firing
rate, kiln exit temperature, kiln rotation rate, scrubber inlet and outlet
temperatures, scrubber water flow rates, and pressure drop across the scrubber
throat.
     The quantity of flue gas produced by the lime kiln is related to the kiln
production rate, the conversion efficiency of CaC03 to CaO, the quantity of fuel
fired in the kiln, and the quantity of excess air allowed into the kiln during
combustion.  The oxygen levels are often monitored at the kiln exit to assist
in the firing of kiln.  In addition, if the kiln is undersized with respect to
the rest of the plant capacity and the kiln becomes the limiting equipment in
the mill production rate, oxygen enrichment of the combustion zone may sometimes
be used to increase the heat release rates in the kiln, which increases the
production rate.  Because all these parameters vary on a daily basis, the
variable-throat venturi scrubber is normally used to control particulate
emissions.  Control of both water flow rates and the throat area usually
results in an unstable controlling operation because of the effect the reaction
and interconnection of the two operating parameters has on pressure drop.  If
a constant pressure drop is maintained across the scrubber regardless of
production rate, the control equipment will maintain its collection efficiency
for smaller particles at both normal and reduced production rates.  At reduced
production rates, the uncontrolled particulate concentration also may be
lower because of reduced .gas velocities through the kiln.  The inspector
should be concerned with any significant change in scrubber pressure drop
and/or any large change in scrubber water flow rates.
     The inspector should check the operation of the scrubber pumps, any
water pressure indicators, and the water flow into any settling pond or tank.
Because not all sources will be equipped with water flow rate indicators,
this check will ensure that' water is flowing into scrubber; however, it will
not ensure that water is being evenly distributed in the scrubber throat or
that the flow rates are correct.
                                     238

-------
     A significant change in both the visible emissions and the values for
pressure drop or water flow rate usually indicates a change in the mechanical
operation of the scrubber.  Examples of mechanical changes might include a
change in controller set point for pressure drop, a change in liquid flow
rate due to a change in valve settings, pump wear, plugged nozzles, or wear
and abrasion of the scrubber throat.  Changes in opacity without changes in
pressure drop or water flow to the scrubber, however, may be related to a
change in process operation or to chemical balances in the scrubber liquid.
This condition would be caused by changes in particle size distribution
rather than by a mechanical change in energy input to the scrubber.  The
additional measurements required to evaluate.the nature of these changes
would constitute a Level III Inspection.
     Level III Inspection—In addition to the aspects of the Level II Inspec-
tion, a Level III Inspection of the lime kiln involves supplemental measure-
ments to assure the collection of correct scrubber data.  The values obtained
during the inspection are compared with those of the design and/or baseline
scrubber parameters to assess scrubber performance.  Some measurements are
difficult to make, however, because of the design aspects of the scrubber.
As a result, the inspector must decide which values should be obtained on the
basis of the type of evaluation to be performed.
     Two important parameters (pressure drop across the scrubber and visible
emissions) have already been discussed.  A change in these parameters may
indicate the need for further investigation.  Although plant instrumentation
should provide a reading of the pressure drop, the inspector should also
measure the pressure drop to confirm the accuracy of this reading.  Measure-
ment of the pressure drop across the scrubber should be taken just before the
converging section and just after the diverging section of the venturi scrubber.
Because a measurement after the cyclone separator will include static pressure
loss associated with mist elimination in the gas stream, the value would not
represent the energy expended in the throat of the venturi scrubber.  If the
design of the scrubber includes a flooded elbow, measurement of the pressure
after the throat may not be possible.  Measurements taken just beyond the
elbow will provide reasonably good data if the inspector allows for about a
1-in. loss associated with the turning of the gas stream at the elbow.
                                     239

-------
     Because the particulate collection efficiency improves with higher
throat velocities  (particularly at smaller particle diameters), the gas
volume through the scrubber is a useful parameter.  Although it is difficult
to know what the effective throat area is during operation, a good estimate
of the range may be calculated for most scrubbers.  Unfortunately, it is
difficult to determine gas volume through a venturi scrubber.  The use of an
F-factor method is not very successful because of the absorption of C02 into
the scrubber liquor.  Many stacks have tangential inlets that introduce
cyclonic flow in the stack, but installing a fan after a cyclone separator
usually destroys any cyclonic flow present after the separator.  Because many
designers use the stack as a final mist eliminator, however, the gas is
introduced tangentially.  Unless flow straightening devices are used, the
cyclonic flow will render velocity traverses by a Pitot tube inaccurate.  The
measured velocity will usually have a positive bias unless the cyclonic flow
is severe, in which case the bias will be negative.  Thus, a Pitot traverse
may not be possible.
     The alternative, the use of fan curves, also may not be acceptable for
several reasons.  First, the gas stream is usually saturated with water
vapor, which decreases gas density.  Second, the presence of C02 and water
droplets increase the effective density of the gas stream.  If it is assumed
that no water droplets are entrained in the gas and that the decrease in gas
density because of water vapor is offset by the presence of C02, the gap must
be treated as if it were air at the scrubber outlet temperature.  This
approach will provide an estimate of gas volume through the scrubber only
if no water is injected into the fan.
     Another reason fan curves may not provide acceptable results is because
of cyclonic flow from the cyclone separator at the inlet of the fan.  Fan
curves and tables are based on smooth transitions and no cyclonic flow into
the fan inlet.  The presence of cyclonic flow at the fan inlet in the direction
of fan rotation will change both the static pressure/volume relationships and
the horsepower/gas volume relationships much the same as a reduction in fan
speed would.  This change occurs because the difference in angular velocities
between the fan rotating speed and the gas rotation speed is reduced.  If the
gas is travelling cyclonically in the same direction as the fan rotation, less
energy is required to accelerate the gas to the rotating discharge velocity

                                     240

-------
of the fan.  The  inspector cannot ascertain the effect on fan performance
without making in-depth tests that are beyond the scope of this guide and
outside the normal inspection procedure.
     If a Pi tot traverse can be performed before the gas enters the scrub-
ber, an estimate  of the gas through the scrubber can be calculated by
assuming that the gas is adiabatically cooled to saturated conditions.  The
procedures for this calculation may be found in any engineering handbook.
Basically, the calculation includes the decrease in actual gas volume due
to cooling and the increase in gas volume due to evaporation of liquid
water to water vapor.  If the volume through the scrubber is assumed to be
the inlet value at the high temperature, the predicted throat velocity
would be too high.  A simplifying assumption is that the gas stream is
cooled and saturated immediately upon entering the scrubber throat and the
resulting gas volume is that which passes through the throat area to yield
an average throat velocity.  If this value is calculated, it should be
compared with design values (if available).   The nominal values for throat
velocity are between 5,000 and 15,000 cm/s,  with values between 10,000 and
15,000 cm/s most common for high-energy venturi scrubbers.
     Another useful parameter in the determination of venturi scrubber per-
formance is the liquid-to-gas ratio.  To find this value, the inspector
must have both the volumetric flow rate of the scrubbing liquor through the
scrubber and the flue gas volume.   He or she must rely on plant instrumen-
tation to determine the liquid flow rate', as no inexpensive portable in-
strumentation is available.  Plant personnel should periodically check the
calibration of this instrumentation to increase the overall  confidence in
the data.  (The usual method is to calculate the ratio between the liquid
volume and the gas volume.)  Typical values  for the lime kiln scrubber are
10 to 20 gallons/1000 acfm of gas.   Higher values usually mean better per-
                                                                         *
formance and higher pressure drop.   Physical measurement of the' liquid
stream is not possible because of the large  quantities involved (between
600 and 1500 gallons per minute).
     As with other scrubbers, good liquid distribution across the throat is
necessary to provide adequate scrubber performance.   Improper coverage be-
cause of nozzle pluggage will allow a portion of the gas stream to bypass the
gas-liquid contact zone and result in a lower pressure drop and lower

                                    241

-------
collection efficiency.  Water flow may remain essentially constant, but the
maldistribution will probably prevent the scrubber from reaching its normal
or maximum pressure drop.  Most larger scrubbers are not equipped with spray
nozzles so as to avoid this problem.  Also, the use of internal overflow
weirs allow the increasing velocity at the throat to atomize the liquid to
droplets.
     Even when pressure drop and water flow rate are at the normally accepted
levels, there are two items that can cause an opacity problem:  total sus-
pended solids and total dissolved solids in the scrubbing liquor.  Because
the kiln is a high-temperature source (exit temperatures of between 400° and
650°F), the resuspension problems associated with high solids levels may be
magnified by this process.
     After scrubbing, the water contains captured particulate that must be
processing further.  Because it is ususally not economical or permissible
to use the water on a once-through basis, the water is recirculated through
the scrubbing system.  One of the advantages of a venturi scrubber is that
it will tolerate much higher levels of suspended solids than other scrubbers
(e.g., an impingement scrubber) because it is less susceptible to pluggage.
Adequate retention time, however, must be provided in the settling or
clarifying tanks for particulate settling and for maintaining the suspended
solids at an acceptable level (15 to 30% by weight).  If excessive levels
of suspended solids are allowed to build up, wear of scrubber components such
as pipes, pumps, and nozzles (if any) will increase.  In addition, the particle
may come out of liquid suspension during vaporization of the liquid in the
scrubber throat and form a particulate that can pass through the scrubber to
the stack.  An equilibrium is eventually established, as the particulate must
exit the scrubber either through the clarifying step or through the stack.
     Some of the solids dissolve and go into solution with the scrubbing
liquid.*  The quantity that goes into solution is limited by the solubility   -
of the various compounds.  The actual level of dissolved solids may be
governed by the solubility limit, but a equilibrium level is usually estab-
lished by the addition of makeup water to account for water losses in the
scrubber.  When the dissolved solids reach a certain level, the water vapori-
zation in the scrubber throat causes the formation of a very fine particle
that produces a visible opacity.  If a change in opacity is noted, water samples
                                     242

-------
should be obtained to determine if the suspended and dissolved solids levels
also have changed.  Although the dissolved solids level is site-specific,
values of 3 to 5 percent or less by weight should prevent the generation of
fine aerosols.
     Part of the problem with solids involves evaporation of the scrubber
liquor and the suspension of particles in the gas stream.  Although the
evaporation cannot be prevented, proper clarifier retention time, adequate
water bleed-off and makeup rates, and possible scrubber changes can signifi-
cantly reduce the opacity problems .that may be associated with the solids
levels.
     If all the parameters compare favorably with the previously established
values (design or baseline test conditions), the performance of the scrubber
is likely to be similar.  Changes in parameters should help identify the prob-
lem or indicate that performance has improved.  The key parameter is the scrub-
ber pressure drop and its relationship to the other operating parameters.

3.5  POWER BOILERS
     In most mills, the kraft recovery boiler could produce a significant
portion of the mill's energy requirements for process steam and electricity.
Because its primary responsibility is chemical recovery, however, the
recovery boiler does not have a very high thermal efficiency.  Therefore,
power boilers are used to supplement the steam and electricity requirements.
Based on a number of economic and regulatory considerations, the number of
boilers and the type of fuel burned varies from mill to mill.  A given mill
may have anywhere from one to eight power boilers burning wood, oil, coal,
natural gas, or a combination of fuels.  This subsection discusses their
potential operating problems along with the major sources of emissions, avail-
able control techniques, potential malfunctions, and inspection techniques
for power boilers and associated control equipment.
     The major emissions from power boilers are particulate matter, S02, NOX,
and C02.  The quantity of each pollutant produced is a function of the fuel
characteristics, the firing method, and the combustion characteristics for
each boiler.  In general, only coal- and bark-fired boilers have particulate
matter control devices.  Although there may be numerous malfunctions that can
                                     243

-------
occur in a power boiler, the two most common operational problems are fuel
quality and establishment of proper excess air levels.
     As with the inspection of the recovery boiler, the inspection of the power
boiler involves the identification and evaluation of those operating param-
eters or variables that indicate operation outside the norm for both the boiler
and the control device.  Section 3.5.5 discusses in detail the major points
to be covered in inspecting both the boiler and the various types of particu-
late matter control devices.  Table 3-27 on page 277 identifies the particular
parameters and variables that should be recorded during a Level III Inspection
of the power boiler.
3.5.1  Process Description
     The power boiler is designed for the efficient production of high-
pressure steam for conversion into electricity through a turbine/generator;
the steam exhausted from the turbine is used for process operation (process
steam heating, operating turbine driven equipment, etc.).  Because steam pro-
duction is the only responsibility of the boiler, it can be designed to
maximize heat transfer and to minimize heat losses through the stack or other
sources.
     Regardless of the fuel, the basic combustion and heat transfer mech-
anisms are the same in the various types of boilers.  Fuel characteristics
dictate such boiler design considerations as furnace volume, heat release
rate, heat transfer surface areas, and tube spacing.  The simplified combus-
tion process involves the mixing of fuel and combustion air in the furnace.
The addition of adequate combustion air, mixing (turbulence), and a suffi-
ciently high temperature cause the fuel to ignite, and the heat generated by
the oxidized fuel is released in the furnace zone.  Some of the heat is used
to sustain combustion, but most of it is used to transfer heat to the boiler
tubes and to generate steam.
     Three heat transfer mechanisms are important to the boiler operation:
radiation, convection, and conduction.  In the furnace zone where the fuel is
combusted, radiation is the dominant heat transfer mechanism.  The heat
absorbed by the tubes is conducted through the tube to the boiler water to
produce steam.  As the combustion flue gas leaves the furnace zone to pass
through the other tube sections, the gas is cooled because of radiant heat
                                     244

-------
loss and the radiant heat transfer decreases.  The gas temperature drops from
2200° to 2400°F in the flame zone down to 1500° to 1600°F before the gas enters
the boiler tubes, where convective heat transfer dominates.  The zones where
the superheater, economizer, and air preheater are placed rely primarily on
convective heat transfer.
     The boiler feedwater enters the economizer under pressure (typically 650
to 1100 psi) to be preheated by the flue gas passing around the economizer
tubes.  The feedwater then enters the steam drum arrangement from which it
passes to the tubes in the furnace walls.  As the preheated feedwater passes
through the tubes in the radiant heat transfer zone, it boils and forms steam.
When the .steam and entrained water droplets leave the furnace walls, they
enter steam drums that separate the water droplets from the steam.  The steam,
which is saturated, is then passed through the superheater to extract more
heat from the combustion gases.  The superheater is needed not only to maintain
boiler efficiency but also to prevent the turbines from operating with a com-
bination of steam and water droplets.  The superheater is located at the
boiler combustion zone outlet, where both radiant and convective heat transfer
occur.
     The steam exiting the superheater is typically limited to a maximum
operating temperature of 1050°F because of the boiler tube metal required to
maintain the discussed pressure ranges.  At most mills, all the superheated
steam is piped to a common header that is fed to one or more turbine/generators
for electrical power generation.  The energy extracted by the turbine reduces
the temperature and the pressure of the steam.  Typical exit pressures are
between 150 and 250 psig to prevent formation of water droplets in other equip-
ment where steam is used (above the saturated steam conditions).  The loop is
completed when the steam is cooled, condensed, and recycled back to the boiler
economizer.
     Not all boilers are equipped with economizers.  Multiple-drum boilers,
which have one or more steam drums and several mud drums, usually allow the
water to boil in the furnace tube walls and  the tubes between the mud drums
and the steam drums.  The steam drums are still used to separate the steam
from the water droplets, and a superheater section is still used to increase
the energy content of the steam.  The thermal efficiency of boilers operating
without economizers, however,  is generally not as high as these equipped with

                                     245

-------
economizers.  As a compromise to reduce the thermal stack losses, most boilers
are equipped with air preheaters that preheat the combustion air before it is
introduced into the furnace; this improves combustion efficiency and increases
the radiant heat released in the boiler furnace zone.  Specific design features
for each type of boiler are discussed.
3.5.1.1  Gas- and Oil-Fired Boilers—
     The design of boilers for firing oil or gas are very nearly identical,
and most boilers can fire either fuel interchangably.  These boilers are
usually physically smaller than boilers having the same production rate but
firing solid fuels because the acceptable heat release rates and velocities
through the tube .banks are higher.  The significant quantities of ash associated
with solid fuels accounts for the differences in physical design.
     The oil/gas boilers used in paper mills can be either shop-assembled
or field-assembled, depending on the steam requirements.  Shop-assembled
"package" boilers are usually selected when additional steam requirements are
small.  These boilers are generally in the 25,000 to 250,000 Ib/h steam range.
They are small, easily shipped by railroad, and relatively inexpensive.  When
larger steam capacity is needed, field-erected boilers in the range of 250,000
to 600,000 Ib/h of steam are usually the norm.  As the name implies, the
boiler components are assembled at the construction site, but in some cases
modular components can be shop assembled to reduce the cost.
     The fuel characteristics and the assembly method affect the design param-
eters for oil/gas fired boilers.  Typical parameters of importance are the
heat release rates (based on furnace volume and radiant heat transfer area),
flue gas velocity through the tubes, and tube spacing, all of which also
affect the physical size of the boiler.  Typical heat release values for oil/
                                o
gas boilers are 200,000 Btu/h/ft  of effective projected radiant heat transfer
                                                                 90
area, regardless of whether the boiler is shop- or field-erected.    The shop-
erected boilers, however, generally have volume-related heat release rates
approximately twice those of field-erected units.  Field-erected boilers are
generally designed with heat release rates of 25,000 to 50,000 Btu/h/ft  of
furnace volume, whereas shop-assembled package boilers will be designed at
                          O On
50,000 to 100,000 Btu/h/ft .    The type of fuel, size limitations, and steam
requirements will affect the value selected.  Natural gas and distillate
                                     246

-------
oils have higher values.  The use of somewhat "dirtier" fuels like No. 6
residual oils will dictate lower values.
     The type of fuel also affects the acceptable velocity through the tubes
and the tube spacing.  Typical velocities through the tubes are approximately
100 ft/s because the flue gas contains very few abrasive particulates.  Tube
spacings are generally 1 to 2 in., depending on the location in the boiler.
Somewhat wider spacings are provided when residual oil is fired.  Draft losses
from the close tube spacing and high velocities are usually controlling factors.
3.5.1.2  Coal-Fired Power Boilers--
     The use of field-erected coal-fired boilers is the alternative to firing
premium fuels (oil and natural gas) in a power boiler.  Coal-fired boilers
are physically larger than their oil/gas counterparts of similar capacity.
They can be stoker-fired or pulverized-coal-fired, depending on the steam re-
quirements.  In most mills, these boilers are either a spreader stoker with
travelling grate or pulverized-coal-fired unit.
     Spreader-stoker boilers are used when a combination of coal and wood is
fired.  Spreader-stoker boilers are limited by grate width and depth, grate
heat release rates, and furnace heat release rates.  Practical stoker boiler
designs limit the size of the boiler to between 50,000 and 400,000 Ib/h steam.
Typically the boilers range in size from 100,000 to 250,000 Ib/h steam.  If
the boiler is coal-fired, up to 450,000 Ib/h is possible (although these are
usually low-pressure, 150 psig steam).
     The physical limitation of most stoker boiler designs is a combination
of grate size and grate heat release rates.  For a spreader-stoker, the
practical limitation for the grate size is approximately 21 ft in depth.  For
higher heat release, however, the furnace must get even wider.  Grate widths
of 32 feet or greater present a physical limitation on the boiler size.  The
grate heat release rates can usually be in the range of 500,000 to 750,000
        0               Of]
Btu/h/ft  of grate area.    Lower values are more conservative and are con-
sistent with unusual design.  The higher value is a maximum, and operation at
this level may cause operational difficulties if coal is not of the correct
quality.  The furnace heat release rates for spreader-stokers are limited to
                         3
25,000 to 32,000 Btu/h/ft  of furnace volume; the lower value is more con-
          90
servative.  .  Coals that have lower ash fusion temperatures will generally re-
quire lower heat release rates.
                                      247

-------
     When the steam capacity of the spreader-stoker is inadequate and coal
quality problems'are expected, or when combination firing of coal with wood
is not planned, the selection of pulverized-coal (p-c) firing may be an
alternative.  From an economic standpoint the break-even point for cost be-
tween stoker and pulverized-coal-firing is in the 200,000 to 250,000 Ib/h
range.  Spreader-stokers rely on burning the coal on the grate, whereas in
p-c firing, the dry, finely ground (70% through 200 mesh) coal is burned in
suspension.
     Although nearly any type of coal can be burned in a p-c boiler, the
boiler must be designed to handle the kind(s) of coal it will normally burn.
Differences in the ash content, the ash fusion temperature, the heat content,
and grindability of the coal will affect the overall design of the boiler.
Unlike oil/gas firing, abrasion and slagging are of concern in both the radiant
furnace zone and the convective boiler passages.  Therefore, p-c boilers are
designed with the aim of preventing slagging.  In general, boilers firing a
Western subbituminous coal with high slagging possibilities will be larger
than a comparable-capacity boiler firing a cleaner coal.  Heat release rates
must be kept below the slagging temperatures, and this requires larger furnace
volumes for a given capacity.
     Three parameters usually help define p-c boiler furnace dimensions:  the
                                                2
volumetric heat release, the heat release per ft  of projected radiant heat
                                                     fi         7
transfer surface, and the plane area heat release (10  Btu/h/ft  of plane area
of the furnace).
coals.
                  Higher heat release values are acceptable for low slagging
                                          3
        Typical values are 22,000 Btu/h/ft  for volumetric heat release rate,
120,000 Btu/h/ft2 of effective radiant transfer area, and 2.1 x 106 Btu/h/ft
of boiler plane area.  Values are lower for coals with high slagging poten-
tial.  Typical values are 15,000 Btu/h/ft3, 70,000 Btu/h/ft2, and 1.4 x 106
        7            Qf) Q1
Btu/h/ft  plane area.  '    These numbers govern the physical size of the
furnace zone.
     The velocity through the convective tube banks in a coal-fired boiler
(typically 50 ft/s) is approximately one-half that in oil/gas boilers.
Tube spacings also are much larger, with spacings ranging from 16 inches in
some superheaters down to about 2 inches in the boiler outlet zones.  Foul-
ing, plugging, and tube abrasion are of some concern when firing solid
fuels, however.
                                     248

-------
3.5.1.3  Wood-Fired Power Boilers--
     Wood-fired boilers, which are normally the travelling-grate type, usually
burn the bark from the debarking drum.  Wood may also be fired in combination
with coal or in a spreader-stoker using gas or oil.  The heat-release limita-
tions for spreader-stokers described in the coal-fired boiler section are
generally applicable to bark boilers.  The bark is generally mass-fed onto a
travelling grate, and the depth of the bark and the grate travel rate control
the steam production rate.  As with a mass-fed travelling grate boiler, the
response to load changes is slow.  Firing of supplemental fuel will aid in
the swing-load capabilities of this kind of boiler.  The boiler load is
usually limited to between 75,000 and 200,000 Ib/h.
3.5.2  Sources of Emissions
     Emissions from all power boilers consist primarily of particulate, S02,
           ?.  The quantity of each pollutant produced is a function of the
fuel characteristics, the firing method, and the combustion characteristics
for each boiler.
NOX, and
3.5.2.1  Gas-Fired Boilers—
     Gas-fired boilers are usually the simplest  boilers from the standpoint
of design and operation.  Emissions generated  by this  "clean fuel" are gen-
erally limited to NO  .  Because  the dominating factor  in  the production  of
                    /\
NO   is the  flame temperature  and the  level  of  excess air,  staged combustion
is used  to  meet excess air requirements  and to reduce  flame temperature  by
reducing turbulence in the combustion zone.  This diffusion-limited  flame is
much more acceptable  in reducing NOV  emissions than  highly turbulent,  high-
                                   X
excess-air  flames.  Low excess air levels  (^ 15%) also promote  higher  boiler
efficiencies.  The emissions  of  CO, hydrocarbons, and  particulate  are
negligible  if complete combustion occurs.
3.5.2.2  Oil-Fired Boilers-
     Fuel oil combustion  generates emissions of  particulate matter,  S02, and
NO  . Although combustion characteristics  are  very similar to  natural  gas fir-
 ing, the potential for pollutant generation is greater.   The  two most  commonly
 used oils are No.  2  distillate  oil and  No.  6 residual  oil.  Normally No. 2
oil  contains less  residual  ash  and sulfur  than the No. 6 oil.
                                      249

-------
      To a certain extent, combustion characteristics control the level of
 participate emissions from oil-fired boilers.  The oil must be atomized and
 mixed with combustion air as it enters the furnace zone.  Improper atomization
 mixing will result in poor combustion and the generation of particulate in
 addition to that present in the oil as a residual ash.  The particulate gen-
 erated as a result of poor combustion will be a combination of fine carbon .  ,
 particles and other unburned hydrocarbons.  Inadequate excess air levels will
 cause similar problems.   Particulate generation rates will otherwise be
 proportional  to the ash  content of the oil.
      As with  gas-fired boilers, NOX emissions from oil-fired boilers are gen-
 erally controlled by combustion modification.  Staged-combustion and diffusion-
 limited flames reduce peak flame temperature and NO  formation by limiting
                                                    A
 excess air.   Highly turbulent flames with high excess air levels promote higher
 levels of NOX emissions.   In addition,  some  oils contain residual  nitrogen
 compounds  that can  tend  to increase NO   production.   In general, tangential
 or corner  firing  of oil-fired boilers produces less  NOX than wall-fired
 boilers.
      The generation  of S02 is generally proportional  to the  level  of sulfur
 in the fuel,  and  most of  the sulfur is  converted to  S02 upon combustion.
 With  sufficiently high temperatures and excess air some of the  S02 will  con-
 vert  to S03 and form sulfuric acid  mist upon  cooling  and contact with water
 vapor.  This  problem is more serious  with  residual oils that contain .vanadium
 (generally 75  ppm or more).   The acid mist forms  very small,  light-scattering
 particles that can cause  a significant  opacity problem.   Acid smuts  may  also
 be produced when  acids condense.on  boiler  heating  surfaces.   Control  of  excess
 air,  switching from  high-sulfur/high-vanadium  oils, or  the use  of an  inhibitor
 to  reduce the  catalytic action  of vanadium on  sulfur  are  the  usual control
 options.
 3.5.2.3  Coal-Fired Boilers—
      Coal-fired boilers produce significant quantities  of particulate,
sulfur oxides, and nitrogen oxides.  The level of each  is related to  the fir-
 ing method, combustion efficiency,  the pollution-control equipment, and the
fuel characteristics.
                                      250

-------
     The level of sulfur oxides produced is generally proportional to the
sulfur content in the coal.  The formation of an acid mist is generally not
a problem unless the sulfur content of the coal is above 2.5 percent.  The
sulfur content of the coal will also affect boiler design and operating
characteristics (sulfur affects the ash fusion temperature) and the choice of
particulate control equipment.  The acid mist problem can be controlled, in
part, by the control of boiler excess air.
     The generation of NOV is strongly related to the combustion method and
                         X
combustion controls.  Free nitrogen in the coal also can be a significant
contributor to overall NO  emissions.
                         A
Most NO  controls deal with the control
       A
of combustion air to the burners (p-c firing).  Staged combustion or off-
stoichiometric firing produces lower peak flame temperatures by limiting the
amount of air available for combustion.  This generally produces longer flames
that are diffusion-limited, which means longer reaction times at lower
temperatures for complete combustion.  Combustion air is provided in excess
of stoichiometric requirements (usually 15 to 30 percent), but the air is
gradually introduced to the flame.   It is very difficult to minimize the
quantity of NOV because of nitrogen in the fuel.
              A
           In general, tangential or
corner-fired p-c boilers will produce less NO  than wall-fired units.
                                             A    -
     Control of NO  in stoker boilers is also a function of controlling both
                  A
overfire and underfire air.  In addition, the proper placement of overfire
air nozzles is critical to the overall efficiency of the boiler.  The operat-
ing principle, however, is the same.  Generally, the underfire air should be
below the stoichiometric rate, which produces a zone in the top of the fuel
bed that is oxygen-deficient and causes distillation and reduction of volatiles
from the bed.  When the underfire gases contact the overfire air at the
high temperatures, combustion is completed.  The overfire air should provide
just enough excess air and turbulence to complete the combustion process
without causing excessive stack heat losses.
     The particulate emission rate  is a function of the coal ash content and
the firing method.  Pulverized-coal boilers will produce more uncontrolled
ash than stoker boilers because the coal is fired in suspension.  Between 70
and 85 percent of the ash in the coal will exit the boiler as fly ash.
Considerably less ash will exit with the flue gas from a stoker-fired boiler
(approximately 30 to 50%).   In addition, the particulate produced by stoker
                                      251

-------
 boilers tends to be much coarser than the ash from p-c boilers  and  will
 generally contain considerably more combustible materials.   These factors
 affect the selection of control  equipment for these boilers.
 3.5.2.4  Bark Boilers—
      The characteristics of bark boilers  are  similar to those of coal-fired
 stoker boilers except with  regard to sulfur oxides emissions.   Because bark
 contains virtually no sulfur,  emissions of S02 are negligible when  bark  is
 fired alone.   Thus,  nitrogen oxides,  particulate,  and unburned  hydrocarbons
 are  the potential  emissions from bark boilers.
      Particulate emissions  from  bark boilers  are relatively  low because  the
 bark is mass-burned  on the  grate rather than  "thrown" on  the grate.  Also,
 the  ash quantities  in bark  can  be significantly less  than those in  some  coal
 supplies.   Thus,  overall  potential  emissions  are less when compared with coal
 firing alone.
      Nitrogen  oxide  emissions potential and control  for bark boilers are
 similar to  those for coal stoker boilers.   More care  is required, however,
 for  proper  adjustment of the overfire and  underfire  air,  because the high
 moisture content of  the  bark (up to  50 percent  by  weight) can cause combus-
 tion problems.   It should be noted  that this  moisture may be beneficial  from
 a NOV  standpoint  because it lowers  the effective flame temperature.  Good
     A
 overfire air distribution and sufficient residence times  in the furnace  zone,
 however,  are needed  for  complete combustion.  This can make control of NO
                                                                         A
 and  particulate emissions very difficult.   In addition, the unburned hydro-
 carbons  will tend to  form very fine,  sticky particulate.  Using natural  gas
 or oil  as a supplemental fuel over  the fuel bed  sometimes minimizes these
 problems  provided boiler design  heat  release  rates are not exceeded.
 3.5.3   Control Techniques
     The  kind of control selected for  power boilers .depends on  fuel type, the
method  of combustion,  fuel  and ash characteristics, and the costs of the
 equipment options available  to meet the prescribed .emission requirements.
 3.5.3.1  Gas-Fired Boilers--
     No add-on control equipment  is applied to natural-gas-fired boilers
 because particulate and S0?  emissions  are negligible  and  NOV is controlled by
                          £               *                 X
 combustion modification.

                                     252

-------
3.5.3.2  Oil-Fired Boilers-
     Depending on the emission requirements, oil-fired boilers may have add-on
particulate control equipment.  Oil-fired boilers that fire No. 2 distillate
oil do not require particulate controls.  Boilers that fire residual oil
efficiently also may not require controls.  If controls are required, they
usually include multicyclones, small ESP's, and scrubbers.  Multicyclones are
generally effective in collecting particles down to 5 to 10 ym in size at 3
to 4 in. pressure drop.  Unfortunately, multicyclones are effective in a very
narrow range of gas volume around the design point.  Small ESP's also can be
used, but the initial capital cost is high compared to that of other control
options.  Scrubbers also may be used, but high operating costs can discourage
this option.  One extra feature that a scrubber can add is its ability to re-
move a portion of the SOp emissions, especially with the use of an alkaline
scrubbing medium.
3.5.3.3  Coal-Fired Boilers--
     Coal-fired boilers require the application of control equipment to meet
particulate emission standards.  Multicyclones, scrubbers, ESP's, or fabric
filters are acceptable control techniques for spreader-stoker boilers.  These
same control options are applicable to p-c fired boilers except that the
multi cyclone would be used only in combination with either an ESP or fabric
filter, not as a separate control device.
     The use of a scrubber for controlling particulate is more practical for
a  spreader-stoker boiler than for a p-c boiler because the stoker boiler pro-
duces  larger particles.  Also, because the mass loading from the spreader
stoker is considerably less, pressure drop and liquid volumes to the scrubber
are  smaller.  The scrubber must be designed to avoid resuspension of particu-
late matter as  a result of the high suspended and dissolved solids in the
scrubbing liquor.  Also, scrubber components must be designed to withstand
possible corrosion.
      Electrostatic precipitators can be used on either stoker or p-c fired
boilers.  Design specifications for the two types will differ considerably,
however.  Stoker boilers generally produce an ash that is composed of carbon
and  other  hydrocarbons, of which 30 to 50 percent is combustible material.
This can  produce a low-resistivity ash that is difficult to retain on the
                                    253

-------
 collection  plates.  Although  the  actual  surface area required to collect the
                                      o
 particulate may  be  low  (100 to  175  ft /1000 acfm) and the power input may be
 high, the ESP must  be designed  with low  superficial velocities to minimize
 particle reentrainment.
     The opposite situation may occur with p-c boilers equipped with ESP's.
 Because combustion  is usually more  complete in a p-c boiler, the other con-
 stituents in the coal play a  major  role  in determining ash characteristics.
 High resistivity can be of major  concern, particularly when low-sulfur/low-
 sodium coals are burned.  If  high resistivities are encountered, the ash tends
 to be very  tenacious when collected in the ESP.  This makes the ash difficult
 to remove from the  plates.  High-resistivity ash causes a significant voltage
 drop across the dust layer, which can cause sparking, back-corona effects,
 and reduced power input to the  ESP.   Potential solutions are to change the
 coal supply; to change operating,temperatures to produce better resistivity
 characteristics; to use such  conditioning agents as ammonia, water, or SO,;
 or to size  the ESP  to operate at  low  power input with adequate treatment
 times.
     Fabric filters are often used  to avoid the problems with resistivity.
 This control technique is suitable  for both stoker and p-c fired boilers.
 Particular attention, however, must be paid to coal quality and firing prac-
 tices of stoker-fired boilers to  avoid bag blinding problems due to unburned
 particles of carbon and sticky  hydrocarbons.  The fabric filter is usually
                                                                        ®
 designed for high-temperature operation, and either fiberglass or Teflon
 bags are used for particulate collection.  The two types of bag cleaning
mechanisms are reverse-air and pulse-jet.  The reverse air mechanism is
                                                               p
 generally limited to an air-to-cloth ratio (A/C) of 2.5 acfm/ft  of cloth
 area.  The pulse-jet mechanism can  operate in a somewhat higher range (A/C
 ratio of 4.0 to 4.5) because of its higher available cleaning energy.   A pre-
 cleaner such as a multicyclone or a simple impaction baffle plate is usually
employed to remove the larger, more-abrasive particulate.   The normal  operat-
 ing pressure drop range for fabric  filter is 3 to 6 in.  of H^O.
     The use of a mechanical  collector or multicyclone as  the sole control de-
vice is generally limited to stoker fired boilers.   The larger particulate and
the lower uncontrolled emission rates make this a suitable control  option  for
some sources.   These devices,  however, operate most effectively at or  near the
design gas volume.   Thus, care must be taken not to exceed the gas  volume.
                                     254

-------
3.5.3.4  Bark Boilers—
     Participate emissions from bark boilers have been controlled by either
mechanical collectors or scrubbers, and sometimes both.  More recently some
bark boilers have been equipped with ESP's to control emissions, but operating
experience is relatively limited.  Although bark boilers behave much as coal-
fired stoker boilers do, a major concern is the generation of fine particulate
as a result of unburned hydrocarbons leaving the furnace.  These particles are
difficult to control when a scrubber or multicyclone is used.  Therefore, com-
bustion control is generally more effective.  These particles are less difficult
to collect in a properly sized ESP, although they are potentially sticky.
This potential stickiness almost precludes the use of fabric filters.
     Variable-throat scrubbers are generally used to maintain a preset pres-
sure drop over a variety of gas flow rates.  Multicyclones are less tolerant
to flow rate changes and perform better with a stable flow near design gas
volume conditions.  Pressure drop requirements for a scrubber may be high  (15
to 25 in. H20), whereas 3 to 4 in. H20 pressure drop is typical for the multi-
cyclone.  The two major disadvantages of scrubbers are the space requirements
and the disposal of the scrubbing liquor containing the collected ash.
3.5.4  Malfunctions
     Numerous malfunctions can occur in power boilers, but the two most com-
mon operational problems are fuel quality and establishment of proper excess
air levels.  Often the excess ai.r level is much higher than needed for com-
plete combustion.  This situation can decrease boiler efficiency, increase
the amount of fuel fired to develop a given quantity of steam, and increase
emissions.
     The presence of the extra nitrogen and oxygen in the excess combustion
air in combination with the combustion gases generally produces a dilution
effect.  Although the peak flame temperature may increase, the average flame
temperature decreases because an extra quantity of gas must be heated.  This
decrease in average temperature decreases the radiant heat transfer to the
furnace walls.  In extreme cases, the extra gas volume will cause incomplete
combustion because of the average decrease in temperature and increase in the
vertical velocity out of the boiler, which causes unburned fuel to be carried
out of the furnace zone.  This is contrary to combustion principles, which
                                      255

-------
 indicate  that a  decrease  in  temperature  requires  an  increase  in  reaction time.
 As  the  flue  gas  leaves  the flame  zone, it  begins  to  lose  heat and starts to
 cool.
   .  Increased gas  volume resulting  from increased excess air means that the
 velocity  of  the  flue  gas  through  the convective tube passes in the boiler must
 increase  because the  available  volume between  the tubes is fixed.  An  increase
 in  velocity  of the  flue gas  over  the tubes will increase  the  convective heat
 transfer  rate.   Unfortunately,  the tube  area is fixed, and the flue gas
 passes  through the  convective heat transfer zone  faster than  normal, and the
 increased heat transfer rate in the  convective section of the boiler never
 quite offsets  the decrease in heat transfer in the radiant zone.  Thus, more
 heat is lost from the boiler during  high excess air conditions than during
 normal  low excess air firing. "The usual indicator of heat loss  is an  increase
 in  stack  temperature and  oxygen levels.  It should be noted that controlling
 the excess air is one way that boiler operators can use to control heat
 transfer  rates in various portions of the boiler.  Improper adjustment or
 failure to calibrate the  controls periodically can reduce boiler efficiency
 and make  steam adjustments difficult.
     Minimum excess air levels differ for the  various fuels,  ranging from 10
 to  15 percent  for natural gas, 10 to  25 percent for oil, 20 to 30 percent for
 p-c boilers,  30  to  50 percent for spreader-stoker boilers, and 75 to 100
 percent for  bark  boilers.  The minimum excess  air levels for  highest boiler
 efficiency are generally  those at which small  levels of carbon monoxide begin
 to  appear.   Boiler  excess air is controlled by the use of C02 or 02 monitors
 at  the outlet  of  the radiant heat zone.  Typically, CO levels are maintained
 at  approximately  100 ppm.  Maximum levels of 400 ppm are usually established
 to  prevent explosions of  CO pockets within the boiler due to  incomplete
 combustion of  the fuel.
     Another boiler problem that shows similar symptoms to high excess air
 levels is the failure of  sootblowers.  Sootblowers may use steam or compressed
air to clean deposits of  the boiler tubes.   The loading and characteristics
of  the ash dictate sootblowing requirements.    Continuous sootblowing may be
required on a p-c boiler, whereas a cycle of once per shift may be sufficient
for oil-fired boilers.  Failure to blow the soot from the boiler tubes allows
deposits to form, which reduces  the heat transfer rate through the tubes.
                                     256

-------
The resulting decrease in efficiency is characterized by an increase in stack
temperature with the 02 or C02 levels remaining the same.
     The mixing of natural gas and combustion air entails relatively few
problems.  With the firing of oil or solid fuel, however, provisions must be
made for effective mixing of combustion air and fuel without excessive turbu-
lence or excessive quantities of excess air.  In oil firing it is necessary
to atomize the fuel in the burner.  This is usually accomplished by using
steam or air at a pressure of at least 30 psig.  The finer the atomization,
the better the combustion, because there is more droplet surface area on
which the combustion processes can take place.  Failure to check routinely
for proper atomizing can result in poor combustion and excessive carbon
carryover.
     In coal-fired boilers, coal quality can be a serious problem.  Problems
vary according to the firing method, but the effect of coal quality is equally
great on emissions and combustion efficiency.  Because of their fundamental
differences, stoker and p-c firing are discussed separately.
     Spreader-stoker boilers are very sensitive to coal characteristics.  For
example, for proper combustion, the size of the coal must be within the range
of 1-1/4 inches to fines.  (See Figure 3-91.)  These size distributions
represent coal delivered to the boiler, not that received from the suppliers.
The spreader-stoker must have some fine coal because 30 to 50 percent of the
coal is burned in suspension and the larger coal falls to the grate to form a
fuel bed.  The mid-sized to larger coal lumps are necessary to form a fuel
bed of the proper porosity for good combustion.  If excessive fines are
present in the coal, too much coal is burned in suspension and carried out of
the boiler, which causes poor combustion (high carbon loss) and excessive
particulate carryover.  Too much handling of the coal can create high levels
of fines.  On the other hand, the lack of fines can have an adverse effect on
combustion because of excessive burning on the grate, unstable flame conditions,
and high carbon loss.  The coal must also be evenly distributed over the
grates from side to side and front to rear.
     Coal sizing is not a big problem in p-c firing except for the maximum
size.  The maximum top size for most pulverizers is about 2 inches.  Very
high quantities of fines may cause some problems with some pulverizers,
but most are easily capable of producing the required fineness (70 to 75
                                     257

-------
95
A ft

oU
/U
CA
OU
^n
40
30
25
O)
5 20
" 15
O)
§ 10
£ 8
1—
•4-> 6
G
 V. \ \ _v
•a 11 -i co AC
-------
percent through a 200-mesh screen).  The fundamental difference between the
p-c boiler and the stoker is that in p-c firing the coal is dried, crushed,
and burned in suspension, whereas in stoker firing it is fired as is.  Sig-
nificant changes in coal characteristics, however, can cause problems.
     Most p-c fired boilers installations have at least one extra pulverizer
to allow routine maintenance to be performed without reducing the boiler
load.  Selecting the number of pulverizers to handle the desired load is
based on the heat content and grindability of the coal because these charac-
teristics define the quantity of coal that must be ground per hour, minute,
day, etc., and the rate at which the coal can be ground to the desired fine-
ness.  They also define the minimum pulverizer capability required and the
necessary redundancy for maintenance.  Because pulverizers are expensive,
excess capacity is held to a minimum.  If the grindability of the coal de-
creases (more difficult to grind) or the heat content of the coal decreases
significantly, or both, the existing pulverizer capability may not be suffic-
ient to provide the necessary fuel input to maintain steam rate.  This is
often indicated by pulverizers operating at full capacity and the boiler still
being unable to reach full load.
     Changes in ash content and other characteristics of the coal can signif-
icantly affect boiler operation.  Increased ash content may increase the
sootblowing requirement and where sootblowing capabilities are marginal, it
can increase boiler stack losses due to inefficient heat transfer.  A more
serious problem is an increase in slagging potential of the ash.  This may
entail derating of the boiler to lower heat release rates to prevent slagging
and increasing the sootblowing requirements.  Boiler operating conditions may
become difficult to control because of a sticky ash coating that reduces heat
transfer efficiency.  These sticky ashes can be very difficult to remove,
and they can cause damage to the boiler tubes due to localized hot spots.
These problems may be the result of changes in a coal seam or the supplying
coal mine.  Coal blending can sometimes lead to similar problems because the
euectic ash formed by the blended coal may have an ash fusion temperature
lower than either coal on its own.
     Improvement in coal quality also can lead to operating problems  in the
p-c boiler.  If the boiler was designed for a slagging coal and a nonslagging
coal is burned, the furnace walls will be too clean for the design conditions.

                                     259

-------
 A higher heat release rate can be tolerated under these conditions,  but too
 much radiant heat is usually adsorbed,  which makes it difficult for,the
 superheater to obtain the high superheated steam temperatures.   Turning off
 or reducing sootblowing helps alleviate this problem.
      Fineness of the coal  from the pulverizers should be checked weekly.
 Failure to feed a fine coal  to the burners may cause incomplete combustion  of
 coal, which allows  unburned,  raw coal to pass out of the boilers to  the
 control  equipment and represents inefficient boiler operation.   Excessively
 fine coal  may result in excessive pulverizing energy being  expended  and the
 possible taxing of  the pulverizer capacity at full  load conditions.   When the
 coal  is  Eastern bituminous,  the test should show that 70 to 75  percent  of the
 pulverized coal  would pass  through a 200-mesh screen.   When the coal  is
 Western  sub-bituminous,  somewhat less fineness (60 to  65 percent)  is  needed
 because  of the "noncaking"  properties of these coals.
     A  tube leak will  cause  boiler shutdown,  and it can affect  the control
 equipment  operations.   Tube  leaks that  are most serious from the standpoint
 of affecting  control  equipment operation are waterwall, boiler  tube,  and
 economizer tube leaks.   Significant quantities of water can escape into the
 flue gas and  cause  possible pluggage of multicyclones  and fabric filters and
 make ash removal  from ESP plates  difficult.   Superheater tube leaks  are not
 as  serious  in  terms  of their  effect on  control  equipment because the  steam  is
 superheated rather  than  saturated.
     In cold weather conditions,  freezing  of the fuel,  particularly  coal
 and wood bark,  can  prohibit proper flow of fuel  to  the  boilers.  The  fuel
 can hang up in  chutes,  hoppers,  or feeders,  and  in  the  case of  coal  shipped
 by rail, can freeze  up  in the  coal  railcars.   On stoker boilers,  hang ups
 in the fuel feed  system  can cause  improper distribution of  fuel  on the
 grates.  When  this occurs, underfire air can  channel to these uncovered
 portions of the grates  because  there is  less  resistance to  air  flow.  This
 reduces the underfire air to other  portions  of the  grate.   This  reduction
may cause distortion of  the grates  because both  the  ash layer and the
 underfire air  help protect the  grates,  the former by providing  an insulat-
 ing layer on the grate surface  and  the  latter  by providing  grate cooling.
 In addition, the channeling of  the  air  destroys  the  air/fuel ratio and  in-
creases emissions.  Residual oil tanks and  fuel  lines are generally heated

                                     260

-------
to maintain the proper oil viscosity.  Failure of these heating systems re-
sults in the inability to pump the oil.
     Each control equipment category has a characteristic group of mal-
functions related to the type of process being controlled and to the design
characteristics of the equipment itself.  The malfunction mechanisms that
have been outlined in previous sections are discussed only briefly.  Mal-
functions in control equipment not used elsewhere in the mill are discussed
in more detail.
3.5.4.1  Mechanical Collectors--
     Mechanical collectors, specifically multicyclones do not have many
applications in the paper mill.  They are generally used to control emissions
from stoker-fired coal, wood, and combination coal/bark-fired boilers.  In a
few cases they are used to control emissions from oil-fired boilers.  They
also can be used as precleaning devices for other control equipment, such as
fabric filters and ESP's.  Because mechanical collectors are relatively
simple to operate, they do not have many failure mechanisms.  Nevertheless,
they require periodic checks and maintenance.  There are several methods in
which multicyclones fail to give the desired performance; some relate to
boiler operation and others are caused by mechanical failure of components.
     Multicyclones are designed to operate in a narrow range around one
design gas volume and uniform particulate density.  When excess air levels
are significantly higher than design, however, the flue gas temperature and
the firing rate are increased to compensate for decreased efficiency.  As a
result, the flue gas volumetric flow rate through the multicyclone can be
substantially higher than design.  This will lead to an increased pressure
drop through the tubes and the application of increased inertia! separation
forces to the particles.  Unfortunately, there is a practical limit to the
amount of pressure drop and inertia! forces that can be applied.  At high gas
volumes, fan capacity may be taxed and the increased pressure drop across the
multicyclone may limit the quantity of gas that may be moved.  When this
"fan-limited" condition exists, the boiler produces more flue gas than the
fan can handle, and the boiler operates under positive static pressure.  Most
boilers are designed to operate under slight negative pressure  (between 0.1
and 0.15 inches H20).  In addition, the inertia! forces applied to the partic-
ulate may be wasted because of a particle "bounce" phenomenon that occurs at
                                     261

-------
very  high  tube velocities.  Operation at the low end of the gas volumes re-
sults  in low  inertia! forces.
     Another  problem lies  in expecting the multicyclone to collect very fine
particulate.  This equipment is not very efficient at collecting particles
below  10 ym in size, which  is one reason why collection of particulate from
p-c boilers by multicyclones alone is generally not acceptable.  Because
stoker-fired  boilers usually produce large particles, multicyclones are very
effective  in  their control.  If combustion problems cause significant quantities
of unburned carbon or hydrocarbons to be produced (as a result of fuel quality
or air distribution problems), however, these particles are likely to pass
through the mechanical collector rather than being collected.
     Fly ash  reinjection is not a satisfactory approach either.  If reinjectipn
is used, the  larger particles (char) with higher carbon contents should be
screened and  reinjected into the boiler, and the fine particulate should be
disposed of.  Reinjection of all of the multicyclone ash means the ash must
either leave  as boiler ash or be reduced down in size until it is capable of
passing through the collector.  Little operating efficiency can be gained by
using this method.
     The material captured by the mechanical collector should not be stored
in the hopper.  Ideally, the ash will be removed by continuous discharge, •
which minimizes the possibility of hopper pluggage.   Hopper pluggage can
cause buckling of the hopper or softening of the tubes due to annealing with
hot ash around the tubes.  Hammering on hoppers until they become deformed
does not help future performance and can actually cause future hopper bridging
problems by establishing a "shelf" in a previously smooth hopper wall.
     If the collector is located on the inlet side of the fan and is under
negative static pressure, the hopper should be sealed by a rotary airlock,
double-gate flapper valves, or some other isolation valve to prevent air
inleakage through the hopper discharge.   Inleakage through the discharge can
cool the ash so that it does not flow as easily.  Inleakage can also resus-
pend the captured ash into the gas stream and destroy or disturb the separa-
tion vortex established in each tube.
     The multicyclone tubes are also capable of plugging.   The tubes can
become plugged in the outlet tube, which will cause the gas entering the tube
to flow out through adjacent tubes and disturb their vortex (Figure 3-92).
                                     262

-------
til
1

§
III
o
o
               Figure 3-92.  Plugged  inlet vane.
                               263

-------
The inlet vanes of the multicyclone can also become plugged and disrupt the
vortex pattern (Figure 3-93).  And finally, the discharge of a tube can
become plugged and allow material to enter the tube and the particulate to be
resuspended (Figure 3-94).  Because approximately 90 percent of the pressure
drop is established across the inlet vanes, a number of tubes could plug
without any significant change in the collector pressure drop.
     Another malfunction mechanism commonly observed in mechanical collectors
is lack of routine inspection and failure to replace vanes and tubes within
the collector.  A maximum period between tube checks should be one year.
These components are subjected to high abrasion forces and cannot perform
their tasks as effectively when they are worn.
     The last malfunction, one that has been encountered occasionally, is
improper sealing between the inlet and outlet of the mechanical collector at
the tube sheet.  Specifically, gasket material used to seal the tubesheet and
tubes must be selected to withstand the normal peak operating temperatures
expected.  Improper selection or installation of the gaskets will result in
the bypass of particulate around the tubes.
3.5.4.2  Scrubbers—
     The use of a scrubber is generally limited to stoker-fired bark and/or
coal boilers, although it can be extended to p-c boilers and oil-fired
boilers.  The operating problems and characteristics are generally similar
to those found on venturi scrubbers applied to lime kilns.  It is suggested
that the section on malfunctions (Section 3.3) be reviewed for a discussion
of scrubber malfunctions.
     Scrubber performance is affected by poor pressure drop maintenance, .im-
proper water flow rates, high suspended and dissolved solids, pluggage and
erosion of pipes and nozzles, pump wear, and to the inability to capture fine
condensable particulate.  These problems, particularly with bark boilers, are
related to incomplete combustion of carbon and hydrocarbons produced in the
fuel bed and are analogous to the evolution of alkali materials from the
kiln.  Unfortunately, these hydrocarbons may not exist as a particulate until
they enter the scrubber throat and are cooled by the contact with the scrub-
bing liquid (evaporative cooling).  In addition, the use of a presaturator to
cool the gas stream and allow particulate growth may enhance particulate col-
lection, but the hydrocarbons may still remain in the very fine particle range
                                    264

-------
Figure. 3-93.  Plugged outlet tube.
              265

-------
                       PLUGGED COLLECTING CONE

                             Dust Discharge
ro
en
         CQ
         fC
         co
(Q
CO
         o
         o
         fD
         O
         3
         UD
         CT
         fD

-------
and pass through the scrubber.  Very high pressure drops may be required to
collect the particulate and prevent a residual blue, haze characteristic of
hydrocarbon emissions.  Proper combustion will also help minimize the problem
of hydrocarbon emissions.                                       ,
3.5.4.3  Fabric Filters-
     Fabric filters are generally applied to p-c boilers and sometimes to
stoker-fired coal boilers.  They are applied to minimize emissions or avoid
resistivity problems associated with some ESP applications.  The capital
costs of fabric filters are generally less than those for ESP's, but operating
costs are usually somewhat higher.
     Failures and malfunctions in a fabric filter can be related to four
general categories:  failure of the cleaning system, failure of the dust
discharge system, failure of the fabric, and a malfunction in the boiler.
Some of the problems are related to two or more areas.  Malfunctions in any
of these areas can affect the overall performance and can cascade to other
problems in the fabric filter.
     An area of major concern when applying fabric filters to stoker-fired
boilers is that of maintaining proper coal properties and combustion in the
boiler.  As has been mentioned previously, the operation of a spreader stoker
is sensitive to the size of the coal, particularly the quantity of fines
in the coal.  If excessive fines are present, the excess carbon carryover can
blind the fabric with a very small, sticky particulate that makes cleaning
difficult.  The result of this blinding is excessively high pressure drops
that can tax the gas-moving capabilities of the fan.  This will eventually
cause a load reduction due to fan-limiting conditions.  Eventually, the fabric
filter will have to be removed from service and the bags replaced.  Most
boilers cannot operate effectively with a pressure drop of 10 to 14 in. H20.
     Another problem related to the high carbon carryover is the potential of
fines in the fabric filter system.  If oxygen levels are high enough in the
presence of the carbon carryover, the result can be the ignition of the car-
bon to a high temperature glow if a spark is carried to the baghouse.  This
                    dD
can destroy a Teflon  bag or take the abrasion-resistant finish off a
fiberglass bag.  If the finish is removed from a fiberglass bag by high
temperature conditions, it will become brittle and abrade itself to form
pinhole leaks and tears in the bag.
                                     267

     A fabric filter may be successfully applied to a spreader-stoker boiler
if combustion air and coal quality are carefully monitored.  Although these
parameters are not problems on p-c fired boilers, similar problems may occur
if the air balance is upset in one of the burners.  The major concern with
these boilers is the temperature of the flue.gas.
     As in other control equipment, dust discharged into the hoppers should
not be stored for any period of time.  A continuous dust discharge from the
hoppers is the best approach.  If allowed to build up in the hopper, the dust
can close off the cleaning contribution of a module and increase the A/C
ratio, pressure drop, and abrasion of the other bags.  Also, hopper outlets
should be sealed against air inleakage to prevent resuspension of the partic-
ulate into the gas stream or excessive cooling of the fabric filter.
     Proper installation of the bags is very important to the overall per-
formance of a fabric filter.  Bags should be the proper material, weight, and
size.  If bags do not meet the basic specifications, they will not survive in
the environment to which they are subjected.  Bag tension and handling of the
bags also affect the bag life.  Improperly tensioned bags or bags that are
mishandled will wear out and develop leaks.
     Some provision should be made for abrasion resistance  in the fabric fil-
ter.  The use of a precleaning device to remove the largest, most abrasive
particulate will generally improve bag life.  The larger particulate  is gen-
erally unable to follow the gas flow and will follow a course that causes it
to impact with the bottom of the bags opposite the inlet.   It is usually the
bottom 18 inches of the bag that has the.worst abrasion damage.  Fabric fil-
ters that have reverse-air cleaning  (where the filtration  is from inside to
outside of the bag), long thimbles, and cuffs on  the bag minimize wear
(Figure 3-95).
     Cleaning system failures can cause significant problems.   Because the
cleaning mechanism governs the maximum A/C ratio  it also controls the cleaning
frequency to some extent.  In general, the higher the  A/C  ratio,  the  more
frequent and more energetic the cleaning energy  requirements.   Failure to
clean the fabric means  an increase in the thickness of the dust layer (which
does most of the filtering) and an increase  in pressure drop.   If prolonged,
this situation can lead to serious long-term consequences.   The response  of
the control systems on  the boiler would be to maintain gas flow rate  by opening
                                      268

-------
                         THIMBLE  AND CLAMP RING DESIGN


                POOR                                    BETTER
CLAMP-
                 SAG
   .IKCREASED
 «r ABRASION
                             SHORT CUFF



rf
«aftr«.... , >
J 	 <--H
.
BAG


1
I
LONG
j THIMBLE ,
J
I
*
> LONG CUFF AND
"^REDUCED ABRASION
^^

eu ri ny / TUBE SHEET '
                POOR
                BAG
           CUFF
           WITH
           SNAP
           R1N6
          J
        "  6AS
            FLOW

       POOR
                                 SNAP  RING  DESIGN
                          SHORT CUFF
                          NO THIHBLE
   INCREASED
„/ABRASION
                             •BETTER
                                                         BAG
                             CUFF WITH
                             SNAP RING
                                                           V6AS I
                                                                        LONG CUFF AND
                                                                      REDUCED ABRASION
TUBE  SHEET
   AND
 THIMBLE
                                    fiAS FLOW
                Figure 3-95.   Fabric  filter bag  attachment methods.
                                           269

-------
 a da'mper.   Thus,  gas flow would likely be the same (for awhile)  at  a  higher
 pressure drop across the fabric.   Unfortunately,  this  pressure drop can mean
 that fine particulate can be drawn further into  the fabric weave (because  of
 the higher available energy drop)  and cause the  particulate  to remain even
 after the cleaning  system is repaired.   This partial blinding of the  bag can
 cause a  permanent increase in pressure drop until  the  bags are replaced.   This
 can happen on both  reverse-air and pulse-jet cleaning  systems, although it is
 usually  more  serious on  pulse-jets.   This problem is usually detected by
 a slight increase in pressure drop if one module  or row is involved and the
 cleaning system fails to activate.  The use of individual manometers  or
 magnehelic gauges on individual  compartments of multicompartmented  baghouses
 will  not detect this problem because  static pressures  and flow will balance
 out as the air will  follow the line of least resistance through  the baghouse.
      Another  problem (limited mostly  to pulse-jet fabric filters) is  the presence
 of moisture and oil  in the compressed air supply.   Air-line  dryers  and oil traps
 should be  provided  to avoid blowing moisture and  oil into the bags  at 90 to 120
 psig.  The moisture and  oil  will combine with the  particulate to  blind the
 bags  and cause a  high pressure drop.   Provisions  for blowdown and other design
 changes  will  help minimize the problem,  but periodic checks  of compressor  rings
 and  seals  in  combination with the  dryer will  virtually eliminate  the  problem.
 3.5.4.4  Electrostatic Precipitators—
      The malfunctions of an  ESP applied to  a power  boiler are generally the
 same  as  those  on  ESP's applied to  the recovery boiler,  with  two  notable excep-
 tions:   1)  fly ash  resistivity must be  considered  in this application, and
 2) ash discharge  designs  for fly ash  are different  than for  salt  cake, although
 problems caused by  the failure of  the discharge system are nearly identical.
      As  discussed in  other sections of  this  guide,  an  ESP charges particles
 through  a  corona  discharge process, and  under the force of an electric field,
moves the  particles  to the collection  plate.  At  the collection  plate, the
 particulate must  leak some of its  charge to  the plate  to complete the electrical
 circuit  in  the ESP.   The  resistivity  of  the  particulate (measured in  ohm-cm)
governs  how easily  the charge is transferred  through the dust layer to the
plate.  The particulate must  be capable  of  retaining some charge so that it
will  remain on the  plate  until it  is  rapped  off.
                                     270.

-------
     The optimum range of particle resistivity  in an  ESP  is  1 x  108 to  1 x
 10   ohm-cm.  This range allows for good collection of  particulate under a
 strong electric field and results in  high particle migration rates.  The par-
 ticles retain enough charge  in this resistivity range to  remain  on the  plates
 under the force of the electric field and to allow electrons to  pass rela-
 tively easily to the plates.  It is when the resistivity  begins  to go outside
 this range that operational  problems  begin.
     Resistivity of particulate is not a constant value;  it depends on  several
 factors, including the chemical composition of  the dust,  the presence of
 moisture and other gases, and the temperature of the  gas.  As Figure 3-96
 shows, resistivity varies over a wide range of  temperatures, and the peak value
 for fly ash resistivity occurs in the 300-350°F range.  Unfortunately,  this
 happens to be the range in which most boilers are operated.  If  the resistivity
 falls outside the optimum range, serious problems may develop.  At low
 temperatures, electrons tend to be conducted over the surface of the particle
 because of the condensation of the conducting species on  the particulate.  At
 high temperatures, the thermal activity of the  electrons  within  the particulate
 allow the electrons to pass through the particulate in a  mechanism called
 bulk conductivity.  At intermediate temperatures, neither mechanism is  very
 effective and the composite effect is an increased resistivity.
     Low resistivity can usually be associated  with high  levels of carbon in
 the particulate.  The particles are charged in  the interelectrode space be-
 tween the discharge electrode and plates, as usual.   During migration to the
 plate, however, the particles leak too much charge to the plate and the
 electric field strength in the dust layer is low.  In the absence of a  strong
 charge,  the particle is free to become resuspended in the gas stream (reen-
 trainment) and to be recollected.   The general  approach to low resistivity is
 to design for low superficial velocities and to minimize  rapping and perhaps
ammonia conditioning to reduce reentrainment potential.    The electric power
 input to the ESP is usually high,  but without the proper  resistivity, per-
formance can be poor.   If low resistivity is caused by poor combustion, com-
bustion problems should be solved.
     The other extreme, high resistivity, is of much more concern in p-c fired
boilers,  where combustion is usually complete.   Fly ash constituents, coal
sulfur content,  and temperature play a major role in  governing the ash
                                 .271

-------
                         RESISITIV1TY
    .13
   10
   10
-  *J1
«  10
ce
   10
    .10
         Minimal
         Lavals,
                     I   I    1
                     Duo to Bulk
                     Conductivity
                     Only
      0   100 200  300 400  600  600

           Gas T«mperatura, °F

      ^_____ Du» to Combinod
            J Bulk and Surface
             Conductivity
Charge*
                                     Putici*
          ». ,-"-•» Surface Conductivity
          ^-N-'
                                                           Bulk Conductivity
             Figure  3-96.  Fly  ash resistivity curve.
                                 272

-------
resistivity.  In general, low-sulfur coals have high potential to cause
operatin problems with high resistivity.
     The high-resistivity condition does not affect the ability of the particle
to accept a charge.  The problems begin when that particle reaches the dust
layer at the plate.  The particles cannot release and transfer the charge
easily, and the charge that is transferred requires considerable energy.  At
the high voltage drops across a relatively thin dust layer, the gas between
the particles begins to break down much like the corona discharge occurring at
the discharge electrode.  Because very high voltage drops are present across
both the dust layer and the interelectrode space, the localized breakdown
causes a spark to occur.  Many of the breakdowns occur through the dust layer,
and high resistivity is characterized by high sparking rates throughout the
ESP.  The critical resistivity is generally around 2 x 10   ohm-cm, where these
localized breakdowns occur, and performance continues to deteriorate as resis-
tivity increases.  During normal operation, random sparking generally does
not cause problems, even though the voltage field collapses during the sparks
and all the power in that field is channelled to the spark.  If sparking is
serious, the T-R controls limit sparking by limiting the strength of the voltage
field and current flow.  This reduces the particle charging rate and the
migration rate in the ESP's.  The voltage and current in the ESP will continue  .
                                                 13
to drop as the resistivity climbs into the 1 x 10   ohm-cm range.
     The sparking, which is a symptom of a phenomenon known as back corona,
finally can no longer take place because operating voltages are too low to
propagate a spark across the interelectrode space.  The voltage drop across
the dust layer, however, is still high enough to cause a breakdown in the
dust layer, and the dust layer produces significant positive particles that
tend to cancel the charging process and effectively reduce the ESP to the
performance of a settling chamber.  Back-corona is usually characterized by
low voltages and high currents, as well as a particular V-I curve relation-
ship.  If the ESP is to operate with any efficiency, it must operate at the
very low levels, below back-corona onset.
     The absence of electrically conducting species such as sodium and sulfur
oxides in the form of sulfuric acid help enhance the resistivity of the fly ash.
The absence of these species increases the potential for problems with high
                                     273

-------
 resistivity, particularly when the ash contains high levels of calcium.  If
 coal supplies vary significantly, resistivity problems may appear periodically.
      The most obvious of several potential solutions is to change the coal
 supply.  Other possible solutions include 1) designing the ESP sufficiently
 large to handle the high-resistivity conditions, 2) adding conditioning agents
 such as water or S03 to the gas stream to enhance the surface conductivity
 characteristics, 3) adding sodium compounds such as sodium carbonate to the
 coal, 4) adjusting the operating temperature to avoid the resistivity problems
 or moving the ESP to a different location (e.g., before the air preheater)
 where temperatures make resistivity more acceptable, or 5) adjusting and
 modifying T-R sets to take advantage of the new pulse-energization technology.
 Each solution must be examined in terms of the site-specific application.
      Bark boiler ash apparently creates no resistivity problems in an ESP.
 Neither does oil-fired boiler ash because there is  sufficient residual  carbon
 in the ash to keep resistivity to acceptable levels.  In the case of oil-fired
 boilers,  however,  there can be a problem with excessive levels of carbon in
 an ESP.   High temperature,  sufficient oxygen,  and sparking can lead to  ESP
 fires,  which can cause catastrophic failure of the  ESP because burning
 temperatures are hot enough to warp the plates and  frame guides.   The fires
 can occur in hoppers,  on  the plates,  or both.   This  is  one reason why ESP's,
 like fabric  filters,  are  usually bypassed or deenergized on  startup.
 3.5.5  Inspection  of Power  Boilers
      This  section  summarizes  the activities  associated  with  power boilers and
 the associated control  equipment.   It  also  identifies  the  data  that should  be
 collected  during a  Level  III  inspection  and  the  procedures  that should  be used
 to  evaluate  these  data.
      The approach  taken during  the  inspection  is to  identify  those  operating
 parameters or variables that  indicate  operation  outside  the norm  for  a  particu-
 lar  boiler/control  equipment  combination.  Normal values or conditions  are
 established  during  the  initial  performance test  or are  based  on the accepted
 state-of-the-art.   This approach  to source evaluation may  be  used  to  determine
 if a more detailed  evaluation or  performance stack test  is required to  verify
compliance with emission standards.
     The following  subsections  summarize  specific areas  that  should be  checked
during the inspection.
                                     274

-------
3.5.5.1  Opacity--
     The inspector should observe opacity according to EPA Reference Method
9.  The observation should be made for at least 30 minutes to determine if
any cyclic patterns are present.  If further evaluation is warranted, the plume
should be observed over a continuous period to identify any "puffing" problems
when ESP's or fabric filters are used.  The latter is a separate activity that
is conducted after the Method 9 observation.  It should be noted that opacity
observations of boilers firing only natural gas should not be necessary,
although a smoking natural gas boiler indicates combustion problems.
3.5.5.2  Transmissometer Data—
     If units are equipped with opacity monitors (transmissometers), the in-
spector should record the current 6-minute average opacity and review the previous
4 hours of monitor output to determine if a cyclic pattern is occurring.
To ensure that the output values are accurate, the inspector should request
the plant to place the monitor in the calibration mode with respect to zero
and span.  As part of the initial monitor certification, the inspector should
have data available on the recorder scale factors and effective stack diameter.
Average opacity readings from the Method 9 observation should be compared with
average transmissometer readings for identical periods.  A major deviation
between the values may indicate possible monitor error.  It should be noted
that the manual and instrument methods are not equivalent.  In sources that
have real-time monitor output, instantaneous opacity spikes will generally
be included in the 6-minute average."  Therefore, the 6-minute averages ob-
tained from the monitor will generally be higher than those obtained by the
manual method because the duration of the spikes are too short to be ob-
served by the inspector.  For ESP's with transmissometers, the inspector should
note the frequency and magnitude of rapper spikes and visually determine if
an inlet-to-outlet field rapper pattern is occurring.  If separate monitors
are installed in each duct, the inspector should evaluate the opacity and
rapper reentrainment pattern in each chamber.  The opacity should be compared
with a typical baseline value for the boiler during known emission periods
(i.e., performance tests).  The opacity data should be used to evaluate ESP
conditions.
     Serious deviations in opacity between chambers can indicate gas flow
maldistribution, an increase in penetration through one chamber as a result
of rapper failure, inleakage, or low power input.
                                     275

-------
     For fabric filters, the inspector should note the frequency and magnitude
of spikes after the cleaning cycles.  These can indicate which compartment or
row may have a leak because some slight rise in opacity will normally occur
after cleaning.  Once the dust cake has reestablished itself over the holes,
the opacity should decrease.
     Opacity monitors are rarely used with scrubbers because of the wet plume
conditions; however, they are used occasionally following multicyclones.
3.5.5.3  Boiler Operating Conditions—
     The inspector should record pertinent boiler operating data to determine
boiler operating conditions during the Level III inspection.  These values may
be used to determine if the boiler is operating at normal production levels
and to compare with the historic baseline data obtained during performance
tests.  Major deviations from normal values should be evaluated with respect
to their effect on control equipment performance and emission levels.
     Boiler operating data that should be measured during a performance test
or an inspection are plant-specific.  Each boiler is usually custom-designed
and erected with a unique instrument and control system package.  The level
of instrumentation is specified by the design engineer and purchaser (plant
engineering department).  Based on the size of the boiler, its cost, and the
experience of the purchaser, the instrument package may range from an extremely
straightforward package to one that is quite complex.  In general, a minimum
amount of instrumentation is necessary for safe operation of the boiler, and
this level of instrumentation will be present in all facilities.  More complex
instrument packages can include an automated computer control system, which
allows the source to optimize combustion and increase the overall efficiency
of the operation.
     Most critical boiler parameters are recorded on continuous strip charts
or circular chart recorders, 'and copies may be obtained after the stack test
(at the end of the day) to provide the necessary documentation.  Most mills
require the boiler operator to record key parameters at set intervals on a
log sheet or in a log book.  The log sheet is typically divided into the fol-
lowing general measurement areas:  fuels, forced air, furnace drafts, gas
temperatures, feedwater, steam, and miscellaneous items  (e.g., pulverizers).
     Table 3-27 lists the items or conditions that must be recorded during
the stack test or Level III inspection.  The list is based on a typical boiler
and would require adjustment for individual installations.
                                     276

-------
        TABLE 3-27.  POWER BOILER OPERATING PARAMETERS TO BE RECORDED
              DURING PERFORMANCE TESTS OR LEVEL III INSPECTIONS
     Parameter
     Variable
  Units
Fuels
Forced air
Furnace drafts
Oil flow
Number of guns
Oil pressure
Oil temperature
Natural gas rate
Coal feed
Primary air flow
Primary air pressure
Primary air temperature
Secondary air flow
Secondary air pressure
Secondary air temperature
Overfire air flow
Overfire air pressure
Damper setting
Underftre air flow
Underfire air pressure
  (each windbox)
Damper setting
Total air flow
Furnace
Boiler inlet
Boiler outlet
Economizer outlet
ID fan inlet
Control equipment inlet
gal/h
None
psig
103 ft3/h
ton/h
scf/min
in. H20
°F
scf/min
in. H20
°F
scf/min
in. H20
                                                                  scf/mi n
                                                                  in.  H20
scf/min
in. H20
in. H20
in. H20
in. H20
in. H20
in. H20
(continued)
                                    277

-------
 TABLE  3-27  (continued)
      Parameter
       Variable
                                                                     Units
 Gas temperatures.
 Feedwater
 Steam
 Miscellaneous
 Boiler outlet
 Economizer outlet
 Evaporator outlet
 ID  fan outlet
 Control  equipment  inlet
 Flow
 Pressure
 Temperature
 Flow
 Drum pressure
 Superheater temperature
 Flue gas oxygen (boiler outlet)
 Pulverizer motor current
 Pulverizer speed
 ID fan motor current
FD fan motor current
                                                                   (l(r Ib/h)
                                                                   psig
                                                                   Or-
                                                                        Ib/h)
                                                                   psig
                                                                  amps
                                                                  rpm
                                                                  amps
                                                                  amps
     Integrator readings also should be recorded at the beginning and ending
of each test run for the following parameters:
     o    Fuel flow (pounds or gallons)
     o    Steam flow (pounds)
     o    Steam used in soot blowing (pounds)
     Based on data obtained from the log, steam tables, integrator readings,
and boiler design data, the following values should be calculated:
     o    Average steam flow for each run
     o    Average fuel  fired for each test run
     o    Heat input (106 Btu/h) to the boiler for each test run for
          each fuel  fired (oil,  natural gas, coal,  wood)
                                    278

-------
      0

      0
        Average boiler output (106 Btu/h) for each test run
        Boiler thermal efficiency (heat output/heat input)
        for each test run
      o    Boiler excess air (percent)
 3.5.5.4  Flue Gas Volume--
      Since control equipment performance is affected by total gas volume, the
 inspector should estimate the volume based on fuel firing rates, flue gas
 oxygen content, and temperature.   Most plants monitor flue gas oxygen at the
 economizer or the furnace outlet for combustion control.   Measurements during
 the Level III should be made at the outlet of the control  equipment to esti-
 mate flue gas volume (except for scrubbers).   The flue gas volume can be cal-
 culated by using an F-factor method and the heat input rate to the boiler.
 The inspector should be equipped  with a portable temperature-measuring device
 (e.g.,  a thermometer or thermocouple) and  portable oxygen-measuring equipment
 (e.g.,  a Fyrite oxygen  analyzer or Orsat).   The contribution of multiple fuels
 may be  determined  if their percent contribution to the overall  heat input rate
 is  known and  the gas  volumes of each  fuel  can  be calculated.   Because all mea-
 surements and calculations  are  at the same  condition,  the  volumes  from combus-
 tion  of each  fuel may be added  to determine the total  gas  volume.   The F-factor
 can be  calculated from  the  fuel analysis, but  standard  tables  are  available
 with  values for  a variety of fuels.   It should  be  noted that oxygen measure-
 ments are made on a dry-basis and  excess air corrections are based on  a  dry-
 basis.
      For  fuels such as coal, oil,  or  natural gas,  the equation  is as  follows:
            Q = 10° Btu/min
                                      20.9
                                  d 20.9-%0,
                                              w
(Tm + 460)
   528
where
  Q =

 Fd =
'fay ~
         gas volume in actual cubic feet per minute, acfm
         dry F-factor for fuel, dscf/106 Btu
         percent oxygen in flue gas
    F  = wet F-factor for fuel, wscf/10  Btu
                                     279

-------
     Tm = measured temperature of flue gas, °F
 The heat input can be calculated if the fuel input rate and the heating value
 of the fuel are known.  For example, for coal:
                106 Btu/min = Ib/min x 106 Btu/lb of coal
                                   or oil:
                106 Btu/min = gal oil/min x 106  Btu/gal
 The factor of (FW - Fd)  accounts for the extra  gas volume due to moisture
 produced during combustion plus  typical  moisture in the fuel.   It must be
 added after correction of the gas volume for excess air because the  02 mea-
 surement is on a dry-basis.
      For bark or wood firing,  the equation becomes somewhat more complex  be-
 cause the moisture percent of these fuels  is very high  and the  contribution
 of the moisture from evaporation to the  flue gas is significantly higher  than
 that produced by combustion.   For wood or  bark,  an equation is  as follows:
            fi          r        9n Q                         T  (T_ + 460)
      Q = 10° Btu/min   Fd x  20 glg0 + (Moisture Correction)      m528	

 This  equation should  be  used when the values  have the same definitions  as above
 and  the  moisture correction  is
Moisture Correction =
      fractional  % H20 in  fuel  by weight x  21.41
(1- fractional  %  H20  in fuel)  x 106 Btu/lb  of dry  fuel
Data on wood and bark usually give weight percent of moisture and heating
value per pound of dry fuel.  The value of 21.41 is the conversion factor
(minus units) to convert 1 Ib of water to 1 Ib of water vapor volume (i.e.,
1 Ib H20 occupies 21.41 scf of volume).
     For scrubbers, measurements should be made at the inlet of the scrubber
and then corrected for the decrease in temperature and increase in water vapor
through the adiabatic cooling of the gas stream.
     When fuel measurements are not available, the boiler heat input rates
may be estimated if the steam conditions, feedwater conditions, and boiler ef-
ficiency are known.  Boiler efficiency can be determined from boiler efficiency

                                    280

-------
charts if temperature and excess air levels are known.  The difference between
heat content of the steam and the feedwater represents an approximate heat
output of the boiler (these values may be determined from standard steam
tables).  Once heat output and boiler efficiency have been calculated, heat
input to the boiler can be calculated as follows:
                    UQ .  T__ll4. _    Heat Output
                    Heat Input - Boiler Efficiency
This value can then be used for the F-factor calculations.
3.5.5.5  Control Equipment Inspections—
     ESP's--The ESP's applied to power boilers are very similar to those of
recovery boilers with respect to their layout and key parameters to be calcu-
lated.  A complete discussion of the data to be taken, the calculations, and
the significance of these parameters were presented earlier.  The major dif-
ference between ESP's on power boilers and recovery boilers is the possibility
of resistivity problems on the power boilers.
     Low resistivities are generally not expected except on stoker-fired
boilers and boilers firing high-sulfur coal.  The electrical signs of operat-
ing problems are high voltage, current, and power input levels that show the
appropriate patterns from the inlet to the outlet of the ESP.   If reentrain-
ment problems are apparent, superficial velocities are normal, and applied
rapping forces are normal, then a low resistivity problem may exist.  Checks
should be made to see if changes have occurred in temperature or fuel quality
that could cause a change in resistivity.
     For ESP's not designed to handle high resistivities, a change in resis-
tivity to higher values can significantly impair performance.   Higher resistivity
may be caused by a change in coal sulfur content, a change in other ash con-
stituents, or a change in the temperature.  Whatever the cause, the performance
usually deteriorates.  As ash becomes more tenacious and difficult to remove
from the plates, power levels decrease and sparking increases throughout the
ESP.  In severe cases, virtually no normally expected increase in power or
current levels occurs from inlet to outlet.  Secondary current levels may just
approach 50 percent of maximum secondary current ratings on the outlet fields
rather than the typical 85 to 100 percent levels.  When fly ash resistivity
has increased to the point where sparking no longer occurs because of a strong
                                     281

-------
 back corona effect, currents will be very high at low voltages and performance
 will be poor.  The voltage-current curve will show distinct negative slopes,
 indicating back-corona.  This condition may be temporary, or it may persist
 for some time if residual ash continues to cling to the plates.
      Fabric Filters—The calculations and measurements that can be made on a
 fabric filter are somewhat limited.   Symptoms of operating and/or design
 problems may appear during the inspection.   For safety reasons, however, an
 internal inspection of the fabric filter is not always possible without
 shutting the boiler down.  Internal  inspections are usually most effective
 for determining  fabric filter problems.
      The first parameter of interest is  the opacity.   Unless a  condensable
 plume is present,  the average opacity should be low.   After each cleaning
 cycle,  opacity will  generally increase slightly as  the bags are placed  back
 in  service with  the dust cake removed.   Some seepage  occurs until  the filter-
 ing action of the  dust cake is  restored.  A significant increase in  opacity
 could indicate a pinhole leak in  a given  module or  a  row of bags.  The  length
 of  time required to  restore* emissions  to  their  previous levels  is  an indication
 of  the  severity of the problem.
      The pressure  drop may give some indication of  the severity of any  holes
 in  the  bags.   The  presence of a few  pinhole  leaks is  unlikely to affect
 the total  pressure drop.   Larger  leaks, however, can  lower  the  pressure drop,
 and complete  failure  of bags  can  cause the  pressure drop  to approach 0.5 to
 1 in.
      An  increase in pressure  drop may  be  the  result of increased gas flow
 through  the fabric filter  or  greater dust layer  resistance  to the  gas flow.
An  F-factor calculation  should indicate any shifts in  gas flow  rate through
 the  fabric filter.  If  no  apparent shift  in gas  volume  has  occurred, blind-
 ing  of the fabric due to failure of the cleaning mechanism  or some other
 related cause may be suspect.  Increased gas volumes through the fabric filter
can  also lead to gradual bag  blinding caused by  the deeper  penetration of the
particulate into the fabric weave, which cleaning cannot remove.  The result
is usually a gradually  increasing pressure drop as the bags slowly become
blinded.  Failure of the cleaning system or injection of moisture into the
fabric filter will usually cause rapid increases in pressure drop.
                                     282

-------
     If no instruments are available to determine pressure drop, static
pressure taps at the inlet and outlet of the fabric filter should provide a
basis for determining pressure drop.  If instruments are available, they
should be checked to see if they are operating properly.  Inlet pressure taps
tend to become plugged because of the dirty gas stream.  The taps also should
be checked to be certain that they are drilled through the walls; occasionally,
this is forgotten during construction.  Sometimes only one tap is connected.
If only the inlet tap is connected, readings may appear to be in the proper
range when they actually are quite erroneous.
     Individual manometers or magnehelic gauges on each compartment generally
will not indicate problems within a specific compartment because the pressure
drop will equalize across compartments.  Because the gas flow through
the compartments will follow the path of least resistance, flow rates through
the various compartments may be unequal, although pressure drops may be the
same.
     The hopper discharge should be checked for plugged or damaged hoppers.
Screw conveyers, rotary airlocks, and other ash removal systems should operate
continuously or at least on a frequent  cycle.  PIuggage of the  hopper can
allow the ash  to build up well into the bags and cause  the bags to be shut
off from the gas flow.  This will increase the effective A/C ratio in the
fabric  filter  and the pressure drop.
     The cleaning system should be  checked for proper  operation, and each
compartment or bag row should be cleaned.  If the  time  between  cleaning  cycles
is  too  long, cleaning mechanism should  be checked  in the manual mode.   Pulse-
jet systems should fire with a resounding thud, with compressed air pressures
of  90  to  120 psig.   Reverse-air systems should isolate each compartment and
the reverse-air and  dwell cycles should be sequenced  to allow  flexing  and
release of the dust  cake  under gentle conditions  (no  "popping"  of  the  bags).
     A more  definitive diagnosis of problems within a  fabric  filter requires
compartment  isolation and  internal  access  to the  baghouse  with the appropriate
safety equipment.  This may be possible during scheduled  outages or, if good
isolation and  purge  systems exist.   Typical  key  points in  an  internal  inspec-
tion include proper  installation  and tensioning  of bags,  the  presence and
patterns of  deposits on  the "clean side"  of the  fabric filter, location and
                                     283

-------
 integrity of the baffle plate, apparent bag/hopper piuggage, moisture or oil
 problems blinding the bag, and evidence of high temperatures in the fabric
 filter.

      Scrubbers—Because the venturi scrubbers applied to power boilers are
 nearly identical to those applied to lime kilns, the reader is referred to
 the section on Lime Kiln Scrubber Inspections for a discussion of items to
 observe or calculate.  There is a difference, however, in the operating pH of
 these two scrubbers.  Lime kiln scrubbers operate under alkaline conditions
 and power boilers operate under acidic conditions.   As a result of the acidic
 conditions,  the power boiler scrubber may have to be more carefully engineered
 to prevent corrosion problems.   Other problems with condensable matter,  solids
 buildups,  and fine particle generation will  be similar for a power boiler.
      Multicyclones—Inspections of multicyclones are relatively limited
 because of restricted access to equipment and the limited number of key  operat-
 ing parameters  to be evaluated.   Other than  checking pressure drop across  the
 multicyclone, checking for proper-hopper  discharge,  and  confirming that  gas
 flow rates are  near  nominal  design  levels, more  detailed checks for proper
 operation  require internal  access to  the multicyclone.   This  will  require
 scheduling a  visit during  a  boiler  outage.
     Multicyclone opacity  levels  usually provide less  information  about  equip-
 ment performance  than  the  opacity levels for  other  pieces  of  control equipment.
 The multicyclone  is  normally unable to  collect the  smaller light scattering
 particles and,  therefore a higher level of opacity  is possible  than with ESP's
 or  fabric filters.   Because the multicyclone  only collects  the  larger sized
 particles, little or no observable shift in opacity may  be  noted even though
 performance has decreased.  As a  result of a  malfunction,  however,  the change
 in  opacity level may indicate a change  in fuel or combustion characteristics
 because these can affect the distribution of  fine particles entering the
multicyclone.
     Pressure drop across the multicyclone is  generally a poor  indicator of
performance and internal multicyclone conditions in the normal pressure drop
range (i.e., 2 to 4  in. H20).  When very low or very high pressure drops are
encountered, it provides an indication that something is wrong  inside the
multicyclone and that maintenance is required.  Small shifts in pressure drop,

                                     284

-------
(<0.5 in.  H20) however, have little meaning in evaluating performance.  Usually
higher pressure drops mean improved performance, but this is dependent on the
initial design parameters.

3.6  OTHER SOURCES
     Several miscellaneous sources at a kraft pulp mill involve the movement
of material and/or the treatment of pulp.   These sources include the bleach
plant and the raw material handling systems.   The major emissions from the
bleach plant are chlorine gas and chlorine dioxide.   The chlorine and chlorine
dioxide are hooded, vented, and passed through a packed bed scrubber using a
caustic solution.  The major malfunctions  at the bleach plant involve the bed
pluggage and channeling and reduction in water flow to the packed bed scrub-
ber.  Because of the toxic nature of the emissions from the bleach plant the
inspection  is generally limited to a review of process and control equipment
specifications.  A checklist to aid in inspecting the bleach plant is provided
in Table 3-31 on page 296.
     The major emissions from the raw material handling system are fugitive
particulate.  In general, the fugitive emissions are controlled by hooding
and venting the emissions to a fabric filter.  Most malfunctions in the raw
material handling area are the result of low capture velocity or fabric filter
failure.  The inspections for the raw material handling area involve a Level
I or Level  II Inspection.  A checklist to  aid in inspecting the raw material
handling system is provided in Table 3-32  on page 299.
3.6.1  Bleach Plant
3.6.1.1  Process Description--
     Due to its nature, the brightness of kraft pulp is low.  In order to use
the pulp for  finer quality white paper, additional lignin must be removed.
The lignin  may be removed through chlorination, oxidation, or reduction.
Chemicals  used in chlorination bleaching are chlorine gas and chlorine dioxide.
     Oxidation may be accomplished by use of hypochlorites, peroxides,
sodium chlorite, and peracetic acid.  Reducing agents include sodium  sulfide,
sodium bisulfide, sodium dithionite, zinc dithionite, and borohydrides.       '
      In the pulp mill, bleaching is accomplished in aqueous solutions under
controlled conditions.  The pulp is bleached and extracted in successive  steps.
After bleaching, the decomposed lignin is removed by alkaline washing.
                                     285

-------
      The following is a discussion of the bleach processes commonly used in
 kraft mills.
      Chlorination—Chlorination is accomplished through the use of chlorine
 gas or chlorine dioxide that is injected into the stock solution and allowed
 to react in a chlorination tower.   The pH of the solution  must be maintained
 below 2.  The retention time is generally between 20 and 60 minutes at  a
 stock concentration of 3 to 4 percent.   Usage rates  are of the order of 3  to
 8 percent of pulp weight.2  Extraction is accomplished  by  alkali  washing in
 successive stages.
      Alkali  extraction—Alkali  is  used to remove residual  bleaching  agents
 and freed lignin.   Bleach products are soluble in alkaline solutions  at  a  pH
 of 10 to 11.   The alkaline solution also dissolves resinous  materials,  pentosons,
 and carbohydrates.   Typical  chemicals  used are hydroxides,  sulfides  of alkali
 metals,  and  calcium hydroxide.
      Extraction  is  generally accomplished at  stock concentrations of  3 to  16
 percent  and  require 60  to 120 minutes  for completion.  The most common extrac-
 tion  chemical  is  sodium hydroxide.   Solution  strength is between  0.5  and 3.0
 percent  on weight of  pulp (air  dried unbleached  tons).2
      Hypochiorites—Hypochiorites  may be  used  in  second or third  stage bleach
 steps  after  chlorination.  Hypochlorite  bleaches  are not specific to  lignin
 and react with the  pulp  cellulose.   Bleaching  is  typically carried out at a
 pH of  6  to 9.5.   Stock  concentration ranges from  12 to 16 percent.  Reten-
 tion time may be  2% hours  for the  first stage and  4 hours for the second
       2
 stage.    The solution pH may  be buffered  by use of caustic soda, sodium
 carbonate, milk of  lime, or  lime.
     Chlorine dioxide—Chlorine dioxide may not be shipped and must be
 generated at the mill site.   It is toxic, corrossive, and explosive.  Above
 50°C it experiences explosive decomposition.   Typical application rates  are
 0.3 to 1.2 percent on weight of pulp at 70°C and at a pH between 3 and 7.2
     Peroxides—Peroxides are generally used as a final  bleach step.  Appli-
cation rates are 0.1. to 0.25 percent on weight of pulp.   Pulp concentrations
are between 12 and 15 percent.  Retention times may be as hfgh as 5 hours at
80 to 85°C and at a pH of'9 to 11.5.
                                    286

-------
     Figure 3-97 shows a typical three-stage bleach line used for sulfate
pulp.  The line consists of a chlorination tower (Clg). chlorination washer,
caustic extraction, caustic washer, hypochlorite tower, and hypochlorite
washer.92'96
     Bleach towers are either up-flow or down-flow.  The height and diameter
are specified to provide the required retention time.  Vacuum washers are
frequently used and are generally covered to reduce chlorine lost to the work
                                               93 97
area.  A vacuum washer is shown in Figure 3-98.  '
     Bleach stages and process are generally identified by use of a shorthand
notation with a series of letters.  Table 3-28 lists the most common de-
signation for each chemical step.  A slash (/) between steps indicates
successive additions of the agents without washing.  Agents placed in paren-
thesis ( ) indicate simultaneous application of agents.  Table 3-29 is a
                                                                   2
list of the most common bleaching sequences used for sulfate pulps.

       TABLE 3-28.  COMMON LETTER DESIGNATIONS USED FOR BLEACH AGENTS2
     A         Acid treatment
     C         Chlorination
     D         Chlorine dioxide
     E         Alkaline extraction
     H         Hypochlorite
     HS        Dithionite
     P         Peroxide
     PA        Peracetic acid
     W         Water soak
     ( )       Simultaneous addition
     /         Successive addition without washing
                                     287

-------
             TABLE 3-29.  BLEACHING SEQUENCES FOR SULFATE PULP
              *CED
              *CEDED
              *CEHH
              *CEHD
               CEDD
               C/HEDD
              *CEH
              *CEHEH
              CC/HEHH
              C(EH)HEH
              CEHHP
              CC/HPH
              CH(EH)D
              CEH(EP)H
              *CEHDP
               CEHDED
              *CEHDP
              *CEHEDP
               CC/HEHEH
               CEHEHH
               CEHEHD
               CEHHDED
               CC/HED/H
               CEHCHDED
              *HCEH
              *CHEH
              *CHED
Bleaching sequences for hardwood sulfate pulp.
                                    288

-------
                                                                                                                       FRESH WATER
00
                  LEGEND

                   -A- REMOTE OPERATED VflLVE
                      GATE VALVE
                      CHECK VALVE
                   -©- FLOW INDICATOR OR CONTROLLER
               i    .% AIR DRY CONSISTENCY
                     *3 CHLOROMIX
                      TOWER
                              Figure 3-97.   Flow  diagram of a three-stage bleach  plant:  CEH.92

       Source:   Reproduced with permission of  Joint  Textbook Committee of the  Paper Industry.

-------
                                                                    93
       Figure 3-98.  Cutaway of a vacuum washer with short drop leg.



Source:  Reproduced with permission of Sandy Hill  Corporation.




                                    290

-------
     The bleaching sequence and chemical type determine the brightness of the
pulp.  Table 3-30 shows the most common sequences and the brightness attain-
able with each sequence.

     .    TABLE 3-30.  BLEACHING SEQUENCES FOR HARDWOOD SULFATE PULP2
Brightness
75
75 to 80
80 to 85
85 to 90
90 plus
Sequences
CEH
CEHH,
CHEH,
CEHD,
CEHEDP
Chlorine dioxide manufacture--Because of

CED, HCEH
CEHEH, CED, CEHDP, CEHD
CHED, CEDED, CEHDP

the explosive nature of chlorin<
dioxide, it cannot be safely liquified or transported.  Mills that use the
material produce the gas on site by one of four commercial methods:  Solvay,
Mathieson, Rapson, and Day Kesting.  The Solvay and Mathieson are the most
common processes.
     The Solvay process reacts sodium chlorate, sulfuric acid, methanol, and
air under pressure to form chlorine dioxide, sodium sulfate, and formic acid
(Figure 3-99).    The gas is adsorbed in a packed-bed scrubber and placed
in storage.  In the Mathieson process, sodium chlorate, sulfur dioxide, and
sulfuric acid are reacted to form chlorine dioxide and sodium sulfate.  The
gas is adsorbed in a dual-stage packed-bed scrubber and placed in storage.
     Chlorine cells—Chlorine is a greenish-yellow gas that is nonexplosive,
nonflammable, very toxic, and a Tung irritant.   The gas is heavier than air
(2.5 times more dense than air) and may collect in pockets where air circu-
lation is poor.
     Chlorine is received at the mill as a dry liquid and is converted to gas
through vaporization for use in bleach towers.   When placed in water, the gas
forms hypochloric and hypochlorous acids.   The acids are extremely corrosive
and at 150°C rapidly attack carbon steel.
                                     291

-------
           COOLER |—I
             COLD
             COLD
                    NO. 1
                    REACTOR
                                JACKET
                                WATER
HOT
H,0
X,
N
R
^
                                                     CHILLED WATER
                                                         ABSORBER
NO. 2
REACTOR
                                                           H,0
    AIR
                                          TO RECOVERY
                                                          CIOSOLUTION
                                                             STORAGE
                                                                        TO PROCESS
               Figure 3-99.   Chlorine  dioxide generating  system.

Source:   Reproduced with  permission of Taylor  Instrument Company.
                                         292

-------
     Many plants produce chlorine on site for use in bleaching.  The most
common process used for manufacture is the electrolysis of a sodium chloride
solution.  The reaction is expressed as follows:
          2 NAC1 + 2 H20 + electric current -»• C12 + H2 + 2 NaOH
     About 12 plants manufacture chlorine by this process.  The two major
electrolysis systems are diaphragm and mercury cell.
     In the diaphram plant, a solution of sodium chloride is passed through
the cell and unreacted salt is resaturated and returned to the reactor.  Spent
liquor from the cell is concentrated in multiple-effect evaporations and
cooled to allow crystallization of the salt.  Caustic is separated and the
salt filtered and washed.   Chlorine released in the reaction is cooled, dryed
with sulfuric acid, compressed and liquified, and placed in storage.  Chlorine
from the reactor has a purity of 97 to 98 percent before cleanup.  Hydrogen
is also recovered from the cell and either oxidized or collected for sale.
     In the mercury cell,  the iron cathode is replaced by a thin layer of
mercury.  Metallic sodium that is formed in the reaction is adsorbed on the
mercury.  Analgam is passed through a packed tower of graphite where it reacts
with water to form sodium hydroxide.  The freed mercury is returned to the
cell.  The hydrogen that is produced is cooled and scrubbed for use as fuel
for the chemical reaction.  The system must be maintained under positive
pressure to prevent air inleakage.
     Hypochlorite manufacture—Many plants produce hypochlorite bleaches on-
site by either batch or continuous process.  Calcium hypochlorite is manu-
factured when calcium hydroxide reacts with chlorine.  Sodium hypochlorite is
formed by the reaction of sodium hydroxide with chlorine.  Sodium chloride is
                                                       98
a waste reaction product and therefore must be removed.
3.6.1.2  Sources of Emission and Control--
     Chlorine gas and chlorine dioxide represent a potential danger in pulp
mills.  The gases are toxic, heavier than air, corrosive, and lung irritants.
Chlorine dioxide gas is also explosive.  In the bleach process, residual
chlorine must be removed through washing in vacuum washers.  The gases in
these systems contain traces of chlorine and chlorine dioxide.  The bleach
tower, washers, and seal boxes are hooded and vented to remove these gases
                                     293

-------
 from the  work place.   The  rate  of ventilation  varies  depending on the tightness
 of the  system and  number of sources  tied  to  each  system.
      Control  is  accomplished by passing the  gas stream  through a packed bed
 scrubber  using a caustic solution.   Typical  superficial velocities through
 the bed are 8 to 10 ft/s,  and the liquor-to-gas ratios  are 3 and 5 gal/1000
 acfm.     The  scrubbers are generally made of fiber reenforced plastics (FRP)
 and are designed for  low temperature.  Under normal conditions, the scrubbers
 can obtain removal efficiencies  of 98 to  99  percent.  Final concentration of
 ClOg in the gas  stream is  a  function of water  temperature and C102 concen-
 tration (g/1)  in the  water.12'99'100
 3.6.1.3  Malfunctions— .
      Bleach plant  emissions  are  controlled by  hooding,  ventilation systems,
 and packed-bed scrubbers.  Malfunctions to the hooding  and ventilation systems
 normally  do not  occur because these  sources  are in work areas.  Loss of hood
 velocity  and/or  ventilation  rates  are usually  quickly corrected because of
 constant  monitoring of the work  atmosphere required for worker protection.
 Malfunctions  in  this  area  are therefore generally associated with the packed-bed
 scrubber.  The most common problems  involve  bed pluggage and channeling, and
 reduction in water flow.   Channeling is typically .caused by bed pluggage as
 a  result  of pulp carry-over  into  the ventilation system or shifting of the
 bed  packing.   Channeling can  also  occur as a result of  poor water distribution
 within  the packed bed.  As a  result  of the channeling,  a disproportionate
 amount  of gas  is passed through  a  reduced bed cross section.  The increased
 velocity  results in water  entrainment and reduced collection efficiency.
 Liquor  carryover can  also  occur  at high liquor-to-gas ratios as a result of
 bed  flooding.
 3.6.1.4   Inspection of Bleach Plants--
      Inspection  in the bleach plant  is limited to process and control equip-
ment specifications (Level  I  or  Level II).   Because of  the toxic nature of the
 emissions, the inspector should avoid breathing the contaminated gas streams
 and use appropriate respirators while working in this area.  Plant safety re-
 gulations must be strictly obeyed.   Failure  to observe  safety requirements
 can result in  damage  to eyes, skin, and lung tissue.
                                     294

-------
     The inspector should document the physical arrangement of the bleach line
 (towers, washers, etc.) and the chemicals used in each step.  Rate of applica-
 tion of each bleach or extraction agent should be documented based on weight
 of pulp.  The pulp tonnage should also be documented.  Points of ventilation
 should be noted, and the rate of ventilation should be determined.
     The design specifications of the packed-bed scrubber should be obtained.
 Specific design variables including cross-sectional area, liquor flow rate,
 superficial velocity, and gas volume should be calculated.
     Specific malfunctions such as pluggage or bed flooding should be deter-
 mined based on the presence of mist carry-over on high scrubber pressure drop.
 High chlorine penetration can be observed as a bluish-green gas emitted from
 the scrubber stack.  Table 3-31 is a checklist for use in the inspection of
 the bleach plant.
 3.6.2  Raw Material Handling Systems
 3.6.2.1  Process Description—           •   ]    .    -
     Many sources in the mill  generate fugitive emissions as a result of
material handling.  These sources are not unique to specific mills, but the
number and magnitude of sources vary from mill to mill.  These sources include
pebble lime silos, hot lime conveying and storage, salt cake unloading and
storage, starch/clay unloading and storage,  salt unloading and storage, and
pigment unloading and storage.
     The rate of emission and process weight of these sources are highly
variable.   The material  can be transferred through the use of bucket elevators,
drag chain conveyors, screw conveyors, or a  pneumatic conveying system.  The
sources may either be contained or open depending on mill design and age.
3.6.2.2  Sources of Emissions  and Control--
     Hot lime conveyor—Hot lime from the lime'kiln burner end hood must be
cooled before being placed in  storage or returned to the slaker.   The lime is
typically cooled on an open drag chain conveyor.   This conveyor,  because of the
presence of dusty lime,  thermal  drafts, and  chain movement,  is a  source of
fugitive particulate.   Mills have controlled the emissions from this area by
containment of the conveyor and hooding and  venting to a fabric filter.  The
filter usually controls  the hot lime elevator and silos.   The filter must be
designed to handle high  temperatures as well as abrasive material.   Typical

                                     295

-------
Stage No.
          TABLE 3-31.  INSPECTION CHECKLIST FOR USE IN BLEACH PLANT
    Chemical rate on weight of pulp
Ventilation points
Control type
Gas volume
Bed cross section
Superficial velocity
Liquor flow rate
Liquor-to-gas ratio _
Pressure drop
Mist carry-over
Visible plume
Plume color
                       acfm
                       ft
                       ft/s
                       gpm
                       gal/1000 acfm
                       in H20
yes
yes
yes
no
no
no
                                     296

-------
bag materials are Nomex  or polyester.  Systems usually are of the pulse-jet
                                                           2
design and operate at air-to-cloth ratios of 4 to 6 acfm/ft .
     Raw material silos—Mills receive a number of raw materials for use in
the pulp and paper departments.  These materials include salt cake, lime,
starch, resin, and pigments.  The materials are received in a dry condition
by either truck or rail and are generally placed in silos.   Silos may also be
used for day storage within the process area.  Pneumatic conveyors are generally
used to transport the products.  Conveying gases that are vented from the silos
can contain considerable amounts of particulates.  Uncontrolled emission rates
can be as high as 20 gr/acfm of vented gas.  The amount of gas vented can be
highly variable depending on process weight and silo size.   Most systems are
between 300 and 500 acfm.  The most common control device is the fabric filter,
which is usually mounted on top of the receiving silo.  The vented gases are
generally between 70 to 100°F and are dry.  Typical air-to-cloth ratios are
                       2
between 4 and 6 acfm/ft .
3.6.2.3  Malfunctions--
     Most malfunctions in the material handling area occur as a result of low
capture velocity or fabric filter failure.  Capture velocity may be reduced
as a result of high-filter pressure drop  (reduced gas volume) or infiltration
of air into the ventilation system between the hood and fabric filter.  In
this case, the filter is operating at design gas volume, but the source capture
velocity is very low.
     Filter failure occurs as a result of the failure of the filter cleaning
system, fabric structure, or dust discharge system.  Cleaning system failures
may occur as the result of failures in:  pulse diaphragms,  timers, solenoids,
or an air compressor.  Water in the compressed-air system commonly results in
diaphragm failure and problems with dust cake release.  Failure to properly
clean the fabric results in high filter pressure drop and reduced ventilation
volume.
     Fabric failure occurs as a result of fabric abrasion,  chemical damage
or temperature excursion.  Most fabric failures occur as pin holes in the
fabric structure.  Holes result from abrasion as the fiber strength is re-
duced.  Particle penetration occurs though pinholes as the gas stream passes
through the orifice.  As a result of the pressure drop across the orifice, the

                                     297

-------
velocity through the opening is very high.   The loss of one bag in the filter
can reduce filter efficiency by several percent.
     Failure of the dust discharge system results in buildup in the hoppers
that eventually causes complete failure of the system.   Dust discharge failure
occurs as a result of component failure such as screw conveyors or air locks.
Inleakage of water through flanges, welds,  and door gaskets causes hopper
buildup and ptuggage.
3.6.2.4  Inspection—
     During a Level III inspection the inspector should determine the process
weight of the product being transferred.  Depending on the ventilation system
design, the inspector should determine the system flow rate and calculate the
collector air-to-cloth ratio.  If the process is a high-temperature source
(hot lime conveyor), the inlet gas temperature should be measured.
     Operation of the filter cleaning system should be checked by listening
for the operation of the compressed-air pulse-cleaning system  (pulse jet col-
lectors).  Visible emission after pulse firing indicates that  a pin hole may
exist  in the fabric for a given row of bags.  The rotary air lock and screw
conveyors should be checked to determine if any  hopper pluggage is occurring.
      If the system can be shut down, the inspector should open and inspect
the clean side of the collector tube sheet to determine if  penetration is
occurring.  Table 3-32 is a checklist  for use in a Level III inspection of
material handling systems using fabric filters.
                                      298

-------
  TABLE 3-32.  INSPECTION CHECKLIST FOR MATERIAL HANDLING SYSTEMS
Source name
Product
Transfer rate
tons/h
Fabric filter type
Cloth area
Gas volume
Inlet gas temp
Air-to-cloth ratio
Pressure drop
Cleaning system operating
ft2
acfm
°F
acfm/f t2
in. HoO
yes no
Rows not cleaning , , ,
Internal inspection
Clean side deposits
Pin holes
Oil /water or bags
Hopper discharge
Type screw
Functioning yes
yes
yes
yes
air lock
no
no
no
no

Hood capture

  Visible emissions
  Hood condition
        yes
Duct work

  Corrosion
  Inleakage
yes
yes
     good
        no
      poor
no
no
                               299

-------
                          REFERENCES FOR SECTION 3
 1.  U.S.  Environmental  Protection Agency.   Atmospheric  Emissions  From  the
     Pulp  and Paper Manufacturing Industry.   EPA-450/1-73-002,  September  1973.
 2.  Joint Textbook Committee of the Paper  Industry.   Pulp  and  Paper Manu-
     facture, Vol.  I.   The Pulping of Wood,  1969.
 3.  Carthage Machine  Company, Inc.
 4.  Koehring-Waterous Ltd.
 5.  Goderhamm Maching Mfg. Co.
 6.  The W. S. Tyler Company.
           t
 7.  ROTEX, Inc.
 8.  Elliott, R. D., and W. H. deMontmorency.  The Transportation  of Pulp-Wood
     Chips by Pipeline, Pulp Paper Res. Inst. Can.  WR Ind. 144, 1963.
 9.  Hawks, R.  In-house Engineering Data.   PEDCo  Environmental, Inc.
10.  Screw Conveyor Corporation.
11.  Link  Belt Limited.
12.  U.S.  Environmental Protection Agency.   Technology Transfer.   Environmental
     Pollution Control, Pulp and Paper Industry, Part I, Air.   EPA-625/7-76-001,
     October 1976.
13.  Kock, P. A.  Treating Kraft Digester Waste Gases.  M.  S.  Thesis,  Chemical
     Engineering Department, Helsinki Technical University, Finland.
     September 12,  1972.
14.  Miller, J. T.   Methods for Managing Batch Digesters.  Manager Pulping
     System Division,  Rader Company, Inc.
15.  Vinnis, V., and T. Kinnula.  A Method to Control  the Turpentine Recovery
     Process of Batch  Kraft Pulp Digesters.   TAPPI Engineering Conference,
     1981.
16.  Hrulfiord, B.  F., and D. F. Wilson.  Turpentine Concentrations in Kraft
     Mill  Condensate Streams.  Pulp and Paper Mag. of Can.   Vol. 74, No.  6,
     June  1973.
                                    300

-------
17.  Kelski, R.  Kraft Mill Odor Abatement by Condensate Stripping and Waste
     Gas Incineration.  M. S. Thesis, Wood Industry Department,  Helsinki
     Technical University, Finland.  September  18, 1969.

18.  Ellerbe, R. W.  Why, Where, and How U.S. Mills Recover Tall Oil Soap.
     Paper Trade Journal, June 25, 1973,

19.  Propst, M.  Improved Techniques for Soap Recovery. . TAPPI Engineering
     Conference 1981.

20.  Edwards, L., and R. Baldus.  Evaluation and Design of Multiple Effect
     Evaporation Systems for Kraft Black Liquor.

21.  Lankenau, H.  G., and J. T. Badyrka.  Multiple Effect Evaporators - Problem
     and Troubleshooting.  TAPPI Engineering Conference 1981.

22.  Nylander, G.   Report on Forest Industry Waste Waters.  Svensk Papperstidning
     67(15):565-572, August 1964.

23.  Leornados, G., D. Kendall, and N. Barnard.  Odor Threshold  Determinations
     of 53 Odorant Chemicals.  JAPCA 19:91-95,  February 1969.

24.  Wilby, F. V.   Variation in Recognition Odor Threshold of a  Panel.  JAPCA
     19:96-100, February 1969.

25.  Backstrom, B., H. Hellstrom, and F. Kommonen.  Purification of Malodorous
     Sulfur Containing Condensates from Turpentine Separation, Digester Blow
     and Spent Liquor Evaporation at the Oy Kaukas Ab, Kraft Mill.  Paperi
     ja Puu 52(3):113-120, 1970.

26.  Morgan, I. P., and F. E. Murray.   A Comparison of Air and Steam Stripping
     as Methods to Reduce Kraft Pulp Mill Odor  and Toxicity from Contaminated
     Condensate.  Pulp and Paper Magazine of Can. 73(5):62-66, May 1972.

27.  Papic, M. M.,  A. D. Mclntyre, and J. G.  Dunsmore.  Stripping of H2S and
     CH3SH from Aqueous Solutions.  Pulp and Paper Magazine of Can.  Vol. 74,
     No. 10, October 1973.

28.  Rowbottom, B., and G. Wheeler.  Stripping-Incineration System Cuts TRS
     Emissions at Cornwall Pulp and Paper Magazine of Can.  Vol. 76, No. 2,
     February 1975.

29.  Collins, T. T.  The Oxidation of Sulfate Black Liquor and Related
     Problems.  TAPPI 38:172A-175A, August 1955.

30.  Trobeck, K. G.  The BT System for Soda and Heat Recovery in Sulfate Pulp
     Mills.  Paper Trade Journal 133(15):40-48, April 20, 1960.

31.  Bealkowsky, H. W., and C. G. Dellaas.   Stabilization of Douglas Fir Kraft
     Black Liquor.   Paper Mill News 74(35):14-22, September 1, 1951.
                                     301

-------
32.  Waltker, J.  E.,  and  H.  P.  Amberg.   Odor Control  in  the Kraft Pulp
     Industry.  Chemical  Engineering  Progress 66:73-80,  March  1970.

33.  Blosser, R.  0.,  and  H.  B.  H.  Cooper.   Survey  of  Black Liquor Oxidation
     Practices  in the Kraft  Industry.   NCASI Atmospheric Pollution Technical
     Bulletin No.  39.   National  Council  of  the Paper  Industry  for Air and
     Stream  Improvement,  Inc.,  New York, New York,  December 1968.

34.  Padfield,  D.  H.   Control of Odor from  Recovery Units by Direct Contact
     Evaporative  Scrubbers with  Oxidized Black Liquor.   TAPPI  56:83-86,
     January 1973.

35.  Van Donkelaar, A.  Air  Quality Controls in a  Bleached Kraft  Mill.   Pulp
     and Paper  Magazine of Can.  69(18):69-73,  September  20, 1968.

36.  Shah, I. S.,  and W.  D.  Stephenson.  Weak Black Liquor Oxidation:   Its
     Operation  and Performance.  TAPPI  51:87A-94A,  September 1968.

37.  Sarkonen,  K.  V,  B. F. Hrutfiord, L. N.  Johonson, and H. S. Gardner.
     Kraft Odor.   TAPPI,  Vol. 53,  No. 5, May 1970.
38.  Perry, J. H.  (ed.).  Chemical  Engineers  Handbook,  3rd  Edition.
     York, McGraw  Hill Book Company, pp.  1585-1586,  1950.
New
39.  Ghisoni, P.  Elimination of Odors  in a Sulfate  Pulp Mill.  TAPPI  37:
     201-205, May 1955.

40.  The Venemark-Design White Liquor Scrubber.  Swedish Patent 226  789,
     Stockholm, Sweden.

41.  Morrison, J. L.  Collection and Combustion of Noncondensable  Digester
     and Evaporator Gases.  TAPPI, Vol. 52, No. 12,  December  1969.

42.  Martin, G. C.  Fiber Carryover With Blow Tank Exhaust.   TAPPI,  Vol. 52,
     No. 12, December 1969.

43.  Douglass, I. B., M. Lee, R. L. Weichman, and L. Price.   Sources of Odor
     in the Kraft Process.  TAPPI, Vol. 52, No. 9,^September  1969.

44.  Berry, L. R.  Black Liquor Scaling in Multiple  Effect  Evaporators.  TAPPI,
     Vol. 49, No. 4, April 1966.

45.  Blosser, R. 0., and H. B. H. Cooper.  Current Practices  in Thermal
     Oxidation of Noncondensable Gases  in the Kraft  Industry.  Atmospheric
     Pollution Technical Bulletin No. 34, NCASI, New York,  New York, 1967.

46.  Babcock and Wilcox. -Steam/Its Generation and Use.  1978.

47.  Passinen, K.  Chemical Composition of Spent Liquors.   Proceedings of the
     Symposium on Recovery of Pulping Chemicals.  Helsinki, Finland, 1968.
                                     302

-------
48.  Control of Atmospheric Emissions  in  the  Wood  Pulping  Industry.   U.S.
     Department of Health, Education,  and Welfare.   Contract  Mo.  CPA  22-
     69-18.  March 1970.

49.  Rydholm, S. A.  Pulping Processes.   Intersciences  Publishers, New York
     1965.                                            •

50.  Personal Communication.  J.  Blue, Babcock and Wilcox  Company.

51.  Thoen, G. N., G. G. DeHaas,  R. G. Tallent, and  A.  S.  Davis.  Effect of
     Combustion Variables on the  Release  of Odorous  Compounds  From a  Kraft
     Recovery Furnace.  TAPPI, Vol. 51, No. 8, August 1968.

52.  Clement, J. L., J. H. Caulter, and S. Suda.   B&W Kraft Recovery  Unit
     Performance Calculations.  TAPPI, Vol. 46, No.  2,  February  1963.

53.  Borg, A., A. Teder, and Bjorn Warnquist.  Inside a Kraft  Recovery Furnace  •
     Studies on the Origins of Sulfur  and Sodium Emission.  TAPPI Environmental
     Conference, 1973.

54.  Bhada, R. K., H. B. Lange, and H. P. Markant.   Air Pollution From Kraft
     Recovery Units - The Effect  of Operational Variables.  TAPPI Environmental
     Conference, 1972.

55.  Bauer, F. W., and R. M. Dorland.  Canadian Journal  of Technology 32:91,
     1954.

56.  Teller, A.-J., and H. R. Amberg.  Considerations in the Design for TRS
     and -Particulate Recovery from Effluents  of Kraft Recovery Furnaces.  TAPPI
     Environmental Conference.

57.  Lange, H. B,, D. P. Pierce,  and J. W. Kisner.   Emissions  From a  Kraft
     Recovery Boiler - The Effects of  Operational Variables.   TAPPI,  Vol. 57,
     No. 7, July 1974.

58.  Hawks, R., and G. Saunders.  Unpublished Data.  PEDCo Environmental, Inc.

59.  Lang, C. J., G.  G. DeHass, J. V.  Gommis, and W. Nelson.   Recovery Furnace
     Operating Parameter Effects  on S02 Emissions.   TAPPI  56:115, June 1973.

60.  Unpublished data, Kopper's Corporation.

61.  Chamberlain, B., E. Lafkrantz, A. Smith, and R. Wostradowski.  Eliminate
     Recovery Furnace H2S Emissions by Controlling CO.   Pulp and  Paper Magazine
     of Canada, Vol.  79, No. 2, February  1978.

62.  Szabo, M. F., and Y. M. Shah.  Inspection Manual for  Evaluation  of
     Electrostatic Precipitator Performance.  EPA-340/1-79-007, March 1981.

63.  Katz, J.  The Art of Electrostatic Precipitation.   Precipitator  Tech-
     nology, Inc. 1981.
                                      303

-------
64.  Hawks, R.  Unpublished data.  PEDCo Environmental, Inc.
65.  Kopper's Corporation, Baltimore, Maryland.
66.  PEDCo Environmental, Inc.  Identification of Parameters That Affect the
     Particulate Emissions From Recovery Boilers.  June 1982.
67.  Personal communication.  Kopper's Corporation, Baltimore, Maryland.
68.  White, H. J.  Electrostatic Precipitation of Flyash, Part I.  Journal
     of the Air Pollution Control Association, January 1977.
69.  LoCicero, P. M., and P. E. Sjolseth.  Operating Experiences With the Ace
     Recovery Furnace Odor Control System.  TAPPI Environmental Conference.
70.  Weinaug, R. J. Jr.  Early Experiences With a B&W Low Odor Recovery
     System.  TAPPI Environmental Conference,  1972.
71.  Hawks', R.  Unpublished data.  PEDCo Environmental, Inc.
72.  Saunders, G.  Unpublished data.  PEDCo Environmental,  Inc.
73.  Personal communication.  Buell Envirotech,  Inc.
74.  A Manual for the Use of Electrostatic Precipitators to  Collect  Flyash
     Particles.  EPA-600/8-80-025, May 1980.
75.  Henderson, J. S.  Final Survey Results for Noncontact  Recovery  Boiler
     Electrostatic Precipitators.  J. E. Sirrine Co.  TAPPI, Vol. 63, No.  12,
     December 1980.
76.  Gooch, V. P.  Low Temperature Corrosion  by Sulfuric Acid  in Power  Plant
     Systems.  Southern Research ESP  Symposium,  February 1971.
77.  Rylands, J. R., and J. R. Jenkinson.  The Acid Dew Point.  Journal  of
     Institute of Fuel, June  1974.
78.  Stockman, L., and A. Tansen.  Gvensk Paperstidn 62:907-914  (1959)
     Abstr. Bull. Inst. Paper Chem.,  30;1164-1165  (1960).   The Paper Industry,
     p. 215, June 1960.
79.  Zarfoss, J. R.  Clean Air From Paper Mill Recovery Boilers Without
     Corrosion.  Environmental Elements  Corporation, Baltimore, Maryland.
     National Association of  Corrosion Engineers,  IGCI.  Atlanta, Georgia,
     1976.
80.  Hawks, R.  Unpublished data.   PEDCo Environmental,  Inc.
81.  Personal communication.  J. Blue, Babcock and Wilcox.
82.  Timmerman, J.  Physio-Chemical Constant  of  Primary  Systems  in  Concentrated
     Systems, Vol.  3,  Intersciences Publishers,  New York,  1960.
                                      304

-------
 83.  Sallack, J.  A.   An Investigation of Explosions in the Sode Smelt Dissolving
     Operation.   Canadian Pulp and Paper Association, Technical Section,
     June 1955.

 84.  Hawks, R.  Unpublished data.  PEDCo Environmental, Inc.
                                                                               •
 85.  Campbell, A. J.   Factors Affecting White Liquor Quality:  Green Liquor
     Concentration,  Drugs Concentration and Lime Dosage.  Pulp and Paper
     Magazine of Canada.  Vol. 82, No. 4, April 1981.

 86.  Kramm, D. J.  Selection and Use of the Rotary Lime Kiln and Its
     Auxilliaries -  11.  Paper Trade Journal, August 21, 1972.

 87.  Dorr-Oliver, Inc.

 88.  Blosser, R.  0.,  A. L. Caron, R. P.. Fisher, M. E. Franklin, W. J.
     Gillespie.   Factors Affecting TRS Emissions From Lime, Kilns.  TAPPI
     Environmental Conference.

 89.  Hawks, R.  Unpublished data.  PEDCo Environmental, Inc.

 90.  Schwieger,  B.  Power Magazine, Vol. 121, No.  2, February 1977.

 91.  Burback, et al.   Combustion Engineering, Power Magazine.  December 1977.

 92.  Three Stage Bleach Plant, Improved Machinery Co.

 93.  Vacuum Washer.   Sandy Hill Corporation.

 94.  Solvay Chlorine Dioxide Generating System.  Taylor Instrument Companies
     and Allied Chemical Company.

 95.  Hawks, R.  Unpublished data.  PEDCo Environmental, Inc.

 96.  Rapon, W. H., C.  B. Anderson, and D. W. Reeve.  The Effluent-Free
     Bleached Kraft Mill, Part VI, Substantial Substitution of C102 for C12
     in the First Stage of Bleaching.  TAPPI Alkaline Pulping Conference
     1975.

 97.  Gall, R. J., H.  D. Partridge, D. J. Josyka, and G. R. Roseman.  Sequential
     Chlorination -  Its Impact on the Environment.  TAPPI Environmental
     Conference,  1975.

 98.  Reeve, D. W., and W. H. Rapson.  The Recovery of Sodium Chloride From
     Bleached Kraft Pulp Mills.  Pulp and Paper Magazine of Canada, Vol. 71,
     No. 13, July 3,  1970.

 99.  Ekono Oy, Helsinki, Finland.

100.  Haller, I.  F.,  and W. W. Northgraves,   TAPPI 38:199, April 1955.
                                      305

-------

-------
                                  SECTION 4
                          COMPLIANCE DETERMINATIONS

     Historically, the level of inspections practiced by State and local
agencies consisted of visible emission evaluations and a quick walkthrough
of the plant.  The purpose of these inspections was to provide a quick deter-
mination of compliance that could be easily documented and defended, and
that would require a minimum of resources.   This level of inspection, previously
referred to as Level I or II, was effective in documenting major occurrences
of noncompliance especially with respect to visible emission standards.
     In general, these inspections are effective screening tools in determining
potential visible emission violations or potential noncompliance with such
permit stipulations as firing rate, process rate, or fuel characteristics
(sulfur, ash, etc.).  Most agencies use the visible emission standard in  lieu
of requiring a stack test to determine compliance with an applicable particulate
emission standard.  As a result of this practice, many sources were unofficially
allowed to operate at particulate emissions levels above the applicable regu-
latory limit (i.e., up to the point of violation of visible emission standards).
In addition, visible emission observations were rarely made when stack tests
were conducted to determine compliance with the particulate standards.
     In many cases, certification stack tests are conducted under optimum
process and control equipment conditions.  These conditions often do not  re-
flect normal day-to-day operating practices or conditions.  A review of stack
tests conducted on recovery boilers indicates a serious -deficiency in recording
process or control equipment parameters that would allow determination of re-
presentative conditions.   For these and other reasons, it is apparent that
the use of visible emissions and annual certification stack tests are not an
effective method of determining the compliance status of many sources in  a
kraft pulp mill.
     Although the above methods ensure a certain level of compliance and
maintenance of control devices at least on a yearly basis, many operating
                                      307

-------
and maintenance problems occur between inspections and stack tests that can
result in serious noncompliance and which have an adverse impact on the am-
bient air quality.  It has been shown that the application of a more compre-
hensive evaluation of process and control equipment can document serious
operation and maintenance problems that in many cases have not been previously
                         2
noted by plant personnel.
     The more comprehensive inspection technique is referred to as a Level  III
Inspection.  This level of inspection helps to ensure continuous compliance
and is based on the application of basic engineering logic to process equip-
ment and control equipment operation.  Level III Inspections require a basic
understanding of the process and control equipment variables or parameters
that influence emissions and a thorough understanding of the factors that
influence control equipment performance.  The inspector is asked to document
a number of operating parameters or variables and compare changes in these
parameters with a known reference point.  This is very similar to the procedures
that are used in evaluating visible emissions.  For example, an increase in
opacity above a given standard can be an indication of an increase in particu-
late emissions.
     Level III Inspections typically rely on more than a single indicator
of performance unless the emission rate is dominated by a single variable.
Major changes in ESP power input or major increases in ESP gas volume are
considered strong indicators of ESP performance.  In most applications a
number of parameters are. used to support the determination of compliance or
noncompliance.  When the parameters are contradictory, a stack test is re-
quired to certify compliance.  If a stack test is conducted, the inspector
must be assured that the test is conducted under the identical conditions
(process and control equipment) that were observed at the time of noncompliance.
     The use of comprehensive inspections requires that the inspector have an
established baseline for each parameter during a period of known compliance.
These parameters are generally established during a performance stack test
or are based on specific design conditions.  Once the baseline  is established,
the inspector can return to the plant at a later date and determine the degree
of continuous compliance by evaluating certain parameters.
     To ensure complete and accurate documentation of parameters or variables
during the stack test, the inspector should observe stack tests on all  sources
                                      308

-------
 within  his  inspection  jurisdiction.   Although  the  primary purpose  for
 observing the  emission test is  to  verify the representativeness  of the  test
 and  that acceptable  testing procedures  are  being followed,  the process  opera-
 tion during the  test is of  critical  interest to the  inspector from several
 aspects.
      During the  initial  compliance test,  the inspector usually can determine
 the  range of process and control equipment  parameters that  the plant  operator
 and  control  equipment  supplier  consider optimum to achieve  compliance with
 the  applicable emission  standards.   This  information is useful not only for
 establishing representative operating conditions during a specific test, but
 is also useful in selecting or  evaluating operating permit  conditions and
 in assisting the inspector  in evaluating  future performance.  The  overall
 process of  establishing  a benchmark  for the process operation is commonly
 referred to  as "baselining."

 4.1   ESTABLISHING A  BASELINE
      Establishing a  baseline involves documenting all pertinent operating
 parameters as they relate to the emission characteristics of the source.
 This  includes both the process  and control equipment parameters.    The base-
 line  provides a fixed point of  operation or a narrow range of operating para-
meters against which other  determinations can be made.   The concurrent emission
 test  provides a documented  emission rate(s) that may be correlated with process
 and control   equipment operating characteristics derived during the test.  The
 baseline test is useful  in conducting subsequent routine inspections.
     The baseline may be used for several purposes.  First, for existing
 sources, baseline values may be obtained prior to a stack test to assist in
establishing representative operating conditions.   The normal range of values
may be recorded during a period prior to a test, and these values may be speci-
fied  in a testing protocol  to establish representative conditions or used as
a starting point in negotiating the testing protocol  with the plant.  Compari-
son of documented compliance test parameters with those specified  in the pro-
tocol helps  to establish whether the process and control  equipment were operat-
 ing at the specified representative conditions.  Second,  for new sources, the
initial  compliance test establishes the operating parameter values that
                                     309

-------
correspond with the measured emission rate.  These values can then be compared
with the design values.  This provides a fixed reference point for comparison
to future operating data.  Third, the values of the baseline parameters pro-
vide data for evaluating routine inspection data.  By knowing the effects of
the various process and control equipment parameters on emissions, one can
make comparisons to evaluate the direction and magnitude of any changes in
performance.  Fourth, documentation of the baseline data will assist in setting
specific ranges on important parameters for possible inclusion in an operating
permit (if required by the agency).  Finally, the baseline test provides a
fixed reference point for comparing long-term performance trends.  Proper
evaluation of the baseline data may assist in the establishment of preventive
maintenance schedules as well as provide an indication of any design or in-
stallation problems.  In addition, the rate at which the normal operating param-
eters may vary from baseline values may assist the agency in scheduling routine
inspections and periodic compliance tests.
     The types of data that should be recorded are dependent on site-specific
factors such as the type of source or process, and the control equipment in-
stalled.  For that reason, the agency observer and the inspector (if different
from the observer) should become aware of the site-specific factors that affect
the emissions, and take steps to obtain that data.  In some instances it will
be difficult to separate and correlate all effects of the variations in the
process and control equipment parameters.  By acquiring as complete an under-
standing as possible of the facility production process and its controls
prior to the stack test, however, the agency observer can frequently identify
key parameters that will have the most influence on emission levels.  It is
strongly recommended that the baselining only focus on those parameters that
have a documented affect on the emission levels rather than on all possible
parameters that might influence the emission levels.  Collection of data that
have no significance can be inefficient and counterproductive.  Considerable
effort can be involved in recording and analyzing all process and control equip-
ment data normally available at a facility.  The inspector must be selective
as to which data to collect.  Also, certain data may be of a proprietary nature
and considered to be confidential business data that will require special
handling and safekeeping.  It is best not to incur this responsibility if it
can be avoided.
                                     310

-------
     Although it 1s suggested that the agency collect all pertinent data, the
agency should.be certain that 1) the data are needed and a change in its value
has an effect on the operation of the source and 2) the data are accurate to
the point that the recorded value has some meaning.  When an operation and
maintenance program or permit to operate is issued based on a range of process
and control equipment parameters, the agency must be able to measure these
parameters.  As a result* maintenance and calibration of the key parameter
instrumentation is needed prior to the performance test.  Additional instru-
mentation may also be needed in some cases.
     During the baseline source test, the agency observer must essentially
conduct an inspection to obtain all the operating parameters for evaluation.
The only major difference between this and any routine inspection is that
emission testing is occurring simultaneously.  Thus, the observer cannot
spend all of the time observing the test because of other responsibilities.
Process and control equipment parameters should be checked throughout the
test.  Data should be obtained for the week previous to and the week after the
stack test to demonstrate representative conditions during the test.  In many
cases two agency inspectors or observers may be needed to observe the baseline
testing.  The services of the field inspector responsible for that facility is
strongly recommended to enhance the field inspector's knowledge and relation-
ship with the facility.
     The use of the baseline for documenting deviations from normal conditions
requires the establishment of a logic system for each process or control device
operating parameter used.  A substantial change in the parameter is evaluated
based on its impact on the overall  emission levels.  For example, an increase
in recovery boiler firing rate would be evaluated because of its impact on the
uncontrolled emission rate from the boiler.
     The technical data provided in previous sections on control device mal-
functions and process conditions are intended to provide the inspector with
the technical .background to evaluate the emission level changes.  The evalua-
tion may be subjective in many instances.  In other cases, sufficient technical
data are available to accurately predict the emission levels.
     The data obtained are usually sufficient to allow the inspector to
negotiate corrective action with respect to the process and control equip-
ment without the expense of conducting a performance stack test.  Many

                                     .311

-------
deficiencies may be corrected as a result of increased or redirected
maintenance activities.  The ability of inspectors to negotiate such correc-
tive action varies from agency to agency.  The inspector must operate within
his agency's guidelines with regard to negotiating compliance agreements or
issuing notices of violation (NOV).  In some cases the corrective action can
be completed before the notices can be drafted and formally issued.
     The following is a summary of key operating parameters that the inspector
may use to evaluate the overall performance of the process/control equipment
for each major piece of process or control equipment.
4.1.1  Recovery Boiler
     The particulate and SCL emission rates from the recovery boiler are inter-
related because the primary method of S02 and TRS control is to convert these
pollutants to sodium sulfate, which increases the particulate emission rate.
The primary parameters that affect particulate emissions are:  firing rate,
primary air rate, excess air, smelt bed temperature, ESP power, ESP superficial
velocity, and flue gas oxygen.   A shift in many of these parameters indicates
an increase in emissions.  In many cases these shifts in parameters can be
used in support of a requirement to take corrective action or to conduct a
performance stack test.  Table 4-1 summarizes the effects of recovery boiler
and ESP operating parameters on particulate and TRS emission rates.
4.1.2  Smelt Tank
     The primary parameters that may be used to determine compliance from the
smelt dissolving tank are connected with the rate of particulate generated and
the condition of the control devices.  Specifically the rate of generation of
particulate is related to smelt rate (i.e., boiler firing rate and reduction
efficiency) and the amount of particle reentrainment.  The condition of the
control device is related to such variables as superficial velocity and water
flow rate.  Table 4-2 summarizes the effects of smelt tank and venturi scrubber
operating parameters on particulate and TRS emission rates.
4.1.3  Lime Kiln
     Uncontrolled particulate emission rates from the kiln are primarily
affected by parameters that affect the superficial velocity through the kiln,
the particle size of the kiln dust, rate of evolution of volatile particulate,

                                   '    312

-------
  TABLE 4-1.  SUMMARY OF THE EFFECTS OF RECOVERY BOILER AND ESP OPERATING
              PARAMETERS ON PARTICULATE AND TRS EMISSION RATES
Parameter
Firing rate
Primary air
Excess air
Smelt bed temperature
ESP power input
ESP superficial ve-
locity
Flue gas oxygen
Primary air tempera-
ture
Visible emissions
Black liquor sulfidity
Change
Increase
Increase
Increase
Increase
Decrease
Increase
Increase
Decrease
Increase
Increase
Effect on parti cul ate
emission rate
Increase
Increase
Increase
Increase
Increase
Increase
Increase
Decrease
Increase
None
Effect on TRS
emission rate
Increase
Decrease
Decrease
Decrease
None
None
Decrease9
Increase
None
Increase
llf  increase  in oxygen  is  a  result of an  increase  in  primary air  volume.
                                     313

-------
   TABLE 4-2.  SUMMARY OF THE EFFECTS OF SMELT TANK AND VENTURI SCRUBBER
        OPERATING PARAMETERS ON PARTICULATE AND TRS EMISSION RATES
Parameter
Firing rate (smelt
rate)
Shatter jet steam rate
Mesh pad superficial
velocity
Mesh pad back flush
rate (caustic)3
Packed bed water flow
(caustic)3
Venturi scrubber water
flow
Pressure drop
Visible emissions
Change
Increase
Increase
Increase

Decrease

Decrease
Decrease
Decrease
Increase
Effect on particulate
emission rate
Increase
Increase
Increase
1
Increase

Increase
Increase
Increase
Increase
Effect on TRS
emission rate
Increase
Increase
None

Increase

Increase
None
None
None
It should be noted that all  smelt vent control  devices do not use caustic.
In some cases only water is  used.
                                    ,314

-------
and the feed rate to the kiln.  The superficial velocity is a function of kiln
firing rate and temperature profile.  The rate of evolution of volatile par-
ticulate is related to slurry feed rate and the amount of soda present in the
slurry.  Parameters affecting the control device are liquid-to-gas ratio,
pressure drop, and particle size^  TRS emissions are related to mud washing
efficiency (% sodium sulfide expressed as Na20), flue gas oxygen, and lime
mud slurry moisture.  The cold end temperature has an effect on TRS emission
levels.  Both the kiln excess air and temperature profile down the kiln in-
fluence the residence time and oxidation rate of TRS compounds where the kiln
is used as a control device.  Where sulfides are present in an insoluble form
and they cannot be removed through washing, lime mud oxidation may be required.
Table 4-3 summarizes the effects of lime kiln and scrubber operating parameters
on particulate and TRS emission rates.
4.1.4  Slaker
     The rate of green liquor and calcium oxide reacted in the slaker has the
strongest effect on uncontrolled particulate emissions.  The amount of heat re-
leased (i.e., steam generated) and the degree of agitation are related to the
reaction rates.  The condition of the scrubber (i.e., water flow rate, liquor
gas ratio, and pressure drop) also affects the emission rate.  Table 4-4 sum-
marizes the effects of slaker and venturi scrubber operating parameters on
particulate emission rates.
4.1.5  Turpentine Condenser and Multiple-Effect Evaporators
     The rate of TRS emissions from multiple-effect evaporators and the
turpentine condenser is primarily a function of noncondensable gas volume and
tail gas condenser final temperature.  The equilibrium vapor pressure of TRS
in the gas stream is also a function of temperature.  Table 4-5 summarizes the
effects of turpentine condenser and multiple-effect evaporator operating
parameters on TRS emission rates.
4.1.6  Blow Tank and Hot Water Accumulator
     The rate of TRS emissions from the hot water accumulator is a function of
digester operation, the condition of the primary and secondary condensers, and
blow gas volume.  The parameters are generally so interrelated that a single
                                      315

-------
TABLE 4-3., SUMMARY OF THE EFFECTS OF LIME KILN AND SCRUBBER OPERATING
           PARAMETERS ON PARTICULATE AND TRS EMISSION RATES
Parameter
Production rate
Firing rate
Kiln temperature
(firing end)
Kiln oxygen
Slurry moisture
Mud sodium sulfide
content (Na20)
Scrubber L/6 ratio
Scrubber pressure drop
Scrubber throat
velocity
Water maldistribution
venturi throat
Visible emissions
Change
Increase
Increase
Increase
Decrease
Increase
Increase
Decrease
Decrease
Decrease

Increase
Effect on parti cul ate
emission rate
Increase
Increase
Increase
Decrease
Decrease
Increase
Increase
Increase
Increase
Increase
Increase
Effect on TRS
emission rate
Increase
None
None
Increase
Increase
Increase
None
None
None
None
None
                                  316

-------
      TABLE  4-4.   SUMMARY  OF  THE  EFFECTS  OF  SLAKER  AND  VENTURI  SCRUBBER
              OPERATING  PARAMETERS  ON PARTICULATE EMISSION  RATE
Parameter
Causticizing rate
Shower water flow
Venturi L/G ratio
Venturi pressure drop
Visible emissions
Venturi throat velocity
Change
Increase
Decrease
Decrease
Decrease
Increase
Decrease
Effect on parti cul ate
emission rate
Increase
Increase
Increase
Increase
Increase
Increase
   TABLE 4-5.   SUMMARY OF THE EFFECTS OF TURPENTINE CONDENSER AND MULTIPLE-
         EFFECT EVAPORATOR OPERATING PARAMETERS ON TRS EMISSION RATE
   Parameter
 Change
                                                              Effect on TRS
                                                              emission rate
Tail gas condenser in-
  let water temperature
Tail gas condenser
  water flow rate
Noncondensable gas
  volume
Vent gas temperature
Increase
Decrease
Increase
Increase
Increase

Increase

Increase
Increase
                                     317

-------
 parameter analysis  is  not effective in  predicting  emissions.   Generally,
 however,  TRS emissions will  increase if the  condensers  are  plugged.

 4.2   CALCULATION  OF EMISSION RATES
      In most cases  the determination of compliance is based on a series of
 primary control performance  indicators  that  may not generally  be quantifiable.
 In specific  applications,  however,  the  inspector may make reasonable estimates
 of emission  rates using  these indicators.  The following methods have been
 demonstrated to be  effective in  calculating  or estimating the  emission rates
 for various  emission sources.
 4.2.1  TRS Sources
     Because the final  temperature  and  pressure of noncondensable gases define
 the partial  vapor pressure of TRS compounds  emitted from the condenser, these
 values taken in conjunction  with gas  flow can be used to calculate the emis-
 sion rate.   As discussed  in  Section  3,  the concentration of TRS gases is
 directly  related to the condenser outlet temperature.   If the  inspector can
 determine the gas volume and  temperature, the emission  rate in pounds per hour
                                                                     o    o
 may be calculated by the product of  concentration  and flow  (i.e., g/m  x m /min
 x 60 min  T 453.6 g/lb).  This  method  can be  applied to  such sources as the
 turpentine vent and multiple  effect  evaporators.
 4.2.2  Emissions from  Recovery Boilers
     The  efficiency of the ESP serving  the recovery boilers can be determined
 by using  variables that define the power input to  the system and the gas
 volume being treated.  The equation  that is  generally applied is a modified
 version of the Deutsch-Anderson equation.    The equation contains a constant
 that must be estimated or calculated from a  previous baseline performance
 test.   A  detailed discussion of the use and  limitations of  the equation are
 provided  in  Section 3.3.  In order to apply  this method, the inspector must
 be able to determine the flue  gas volume passing through the ESP and the
power input  to the ESP.  The inspector must also be able to determine that the
 unit is in reasonably good operating condition (no gross gas maldistribution,
power distribution imbalances, high resuspension rates, or  high rapper re-
entrainment) because certain occurrences can cause the Deutsch-Anderson equa-
tion to over-predict efficiency.
                                     318

-------
     The value to be used in the equation is determined from previous base-
line tests or it may be estimated.  Typical values for k range from 0.1 to
0.25.5  It is usually advantageous to request that stack tests be performed
over a range of boiler loads (gas volume) and at several ESP power input
levels.  This allows the inspector to determine if the value of k remains
reasonably constant over the range of normal operating conditions.  The use
of the Matts-Ohnfeldt version of the efficiency equation has been shown to
provide a better correlation where units are designed for high specific corona
power levels (> 500 W/1000 acfm).
     Figure 4-1 shows a method of calculating ESP efficiency based on migra-
tion velocity and plate area.  This method is useful in reviewing design
capabilities of an ESP but may not be accurate for evaluating a unit that is
currently in service.  This method is limited because the migration velocity
is not constant over the operating range of the unit.  As the power input to
the unit is decreased, the migration velocity is reduced.  The lower section
of Figure 4-1 shows the method to be used in calculating emissions using ESP
power input and gas volume.  Power input is calculated from secondary meter
readings.  The gas volume is calculated from a modified F-factor or measured
with a Pitot tube.  The most serious deficiency in applying this method is the
determination of the uncontrolled emission rate from the boiler.  As stated
previously, the uncontrolled particulate emission rate is not constant and
is a function of the boiler combustion characteristics.  The emission rate is
a compromise between particulate as sodium sulfate and S02-  If the level
of S02 can be determined to be fixed over the operating range of  interest,
the particulate emission rate may b* assumed to be reasonably constant (Sec-
tion 3.3.1).  If inlet performance stack tests are available, they should be
used in the mass emission rate calculation.
4.2.3  Power Boilers
     Emission rates from power boilers using an ESP may be calculated in a
similar manner to the one used for recovery boilers.  The restrictions and
cautions are the same in applying the equations.  In general, the values of  k
and migration velocity are different from  those applied to recovery boilers
(Section 3.5).
                                      319

-------
Given:
 Boiler:
 ESP:
              Manufacturer B-W
              Indirect contact
              1200 ADTP/day
              Firing rate 350 gpm at 61.5% BLS (149,820 Ib/h BLS)

              Kopper's
              Wet bottom
              2 chambers
              3 fields               „
              Plate area = 151,200 ft
              Superficial velocity = 3.11 ft/s

              SCA = 386 ft2/ 1000 acfm
              Gas volume = 392,000 acfm (6533.3 acfs)

              Gas temperature = 415°F
              Migration velocity =8.0 cm/s (0.262 ft/s)
              Uncontrolled emission rate =3.2 gr/acfm
Calculate:
     1.   Allowable emission (AL)

          AL = 3 lb/3000 Ib BLS
                             = 149.821b/h
     2.    Uncontrolled Emission Rate (UN) based on inlet loading

          UN = (392.000 acfm) (3. 2 gr/acfm) _ fin  .
                        7000 gr/lb           bu min
3.
          UN = 10,752 Ib/h

          Actual  Emission per Deutch-Anderson  equation  for ESP  efficiency
          (AE)   '

          a.    Design efficiency for ESP per Deutch-Anderson  equation:
              (1 -
                              100
              where   n  =  collection  efficiency,  %
                      w  =  migration velocity,  ft/s
                      A  =  electrode collecting area, ft
                      v  =  gas  volume, acfs
     Figure 4-1.  Calculation of  kraft  recovery boiler ESP efficiency.
                                      320

-------
b.
                         -[0.26
               n = (1 - e

               n = (1 - e"6'017) 100
               n = (1 - 0.002437) 100
               n = 99.75%
              AE = UN x (100 - n)/100
                                     ) 100
              AE = (10,752)
              AE = 26.20 Ib/h
          Actual Efficiency for ESP using Modified Deutsch-Anderson equation
               n = 100 [1 - e-°-06 k (p/C»]
               given:  power input 135,000 watts
                       gas volume  392,000 acfm
                       k = 0.135 from previous stack tests
              f  -    n fte/n'iocwl35,000x1
          n =  i .  e-0.06(0.135)(   3g2  )J
          n =
          n =
                    1 - e
                    -3.099
100
                    1 - 0.04501 -100
               n = 95.49%
              AE = UN x (100 - collection efficiency)/100
              AE - (10,752) (
              AE = 484.6 Ib/h

Figure 4-1.  Calculation of kraft recovery boiler ESP efficiency.   (Continued)
                                 321

-------
 4.3  STACK TEST METHODS

      Each State or local agency has adopted stack test requirements that may
 be used to demonstrate compliance with parti oilate, TRS, S02, and visible
 emission standards.  The following is a description of the federally approved
 reference methods.
 4.3.1  Particulate Sampling
      Particulate sampling is generally accomplished through the use of
 Reference Method 5 or 17.  Method 17 uses isokinetic sampling with mass weight
 determined by filter catch.   The filter is at gas stream equilibrium tempera-
 ture and moisture conditions within the source stack.   Method 5 requires
 extractive gas filtering with the gas stream passing the filter to be no
 greater than 250°F + 48°F.   Both methods  use isokinetic sampling methods
 with gas moisture determination by condensation in  the impingers and adsorption
 in the silica gel.   Sample  volume is  measured on a  dry basis  with a  dry gas
 meter.
      The use of an  impinger  catch  in  calculating the mass emission rates vary
 from state to state  with  several  States requiring back-half inclusion  and
 some only  counting material  in  the  first  impinger.
      For most direct-contact recovery  boilers,  the  gas  stream temperature is
 close to the  required temperature for  Method  5  filters, and generally  there
 is  reasonable agreement between Methods 5  and  17.   In  noncontact systems the
 gas  temperature is higher (>425°F)  and the gas  stream must  be cooled in
 Method  5 sampling.  Because  of  the  presence of  condensables,  the Method  5
 catch may be  higher than  a comparable Method  17 catch.
      In  order  to account  for the higher gas temperature and to adjust  the filter
 catch, Method  17 requires the addition of  0.004 gr/dscf to  the measured  value.
A maximum stack temperature  of  400°F should also  be specified.
     The inclusion of a back-half catch results in an increase in  the measured
mass emission  rate.  A portion  of this catch is the result of condensable
particulate passing through  the filter, but in  some cases the weight may be
due to artifact formation.
     In general, there are no technical problems  in applying Method 5 or 17
to any particulate-emitting source in the mill.  Strict adherence to the
                                    322

-------
sample methods Is required, however, to give reproducible and enforceable
results.
     The most serious deficiencies in the sampling protocol of the pulp mill
stack tests fall into two areas 1) failure to fully document process conditions
as required under 40 CFR 60, Appendix A, to prove representativeness of the
test conditions and 2) sampling of scrubber stacks where severe tangential flow
is occurring.  Item one has been discussed in other sections of this manual
and does not require further comment.
     In regard to item two, most kraft mills use low-energy dynamic or venturi
scrubbers that use cyclone separators to remove water droplets from the gas
stream.  The separator creates a highly tangential flow pattern that almost
without exception cannot be sampled in accordance with the requirement of EPA
Reference Methods 1 and 2.  Straightening vanes or other methods must be used
to reduce the tangential flow to acceptable levels to complete the test.
Reference Methods 5 and 17 are provided in Appendix B.
4.3.2  TRS Sampling                                                         •
     TRS sampling is generally accomplished through Methods 16 and 16A.  Method
16 is a semicontinuous method that determines sulfur compound emissions using
the principle of gas chromatographic separation and flame photometric detec-
tion.  The gas sample is extracted from the gas stream and diluted with clean
dry air.  A portion of the sample is analyzed for hydrogen sulfide (H2S),
methyl mercapton (MeSH), dimethyl sulfide (DMS), and dimethyl disulfide (DMDS).
The total of these compounds is defined as TRS.  Moisture, carbon monoxide,
carbon dioxide, particulates, and S02 are interferences for this method and
must be quantified or removed from the gas stream before analysis.  Method
16 is provided in Appendix C for reference.
     Method 16A is the preferred method for TRS sampling because of its
simplicity.  The gas sample is extracted from the stack, and S02 is selectively
removed from the sample by use of a citrate buffer solution.  The gas stream
is then oxidized and analyzed as S02 using a barium-thorin titration procedure.
The analysis for S02 is identical to that used in Reference Method 6.  Method
16A is also provided in Appendix C.
                                      323

-------
4.3.3  SO2 Sampling
     Sampling for S02 is generally accomplished through use of a wet chemistry
method that selectively adsorbs and reacts with S02.  The Reference Method
is EPA Method 6.  A gas sample is extracted from the gas stream, and the S02
and sulfuric acid mist are separated.  The S02 is measured by the barium-
thorin titration method.  Interferences with the method are free ammonia,
water soluble cations, and fluorides.  Cations and fluorides are removed by a
glass wool filter and an isopropanol bubbler.  Ammonia interferes by forming
a particulate sulfide that reacts with the indicator.  The presence of ammonia
can be detected by observing a white particulate precipitant in the isopropanol
bubbler.  Method 6 is provided in Appendix D.
4.3.4  Visible Emissions
     Compliance with visible emission standards is usually documented through
the use of Method 9.  The opacity is determined visually by a qualified
observer.  In general, the observer visually determines the opacity of an
emitted plume at 15-second intervals for a period of 6 minutes or as required
by agency policy.  The averaging time for the observations in Method 9 is 6
minutes.  The averaging time varies from state to state, and an exclusion
period above the standard is usually permitted.  The presence of uncombined
or condensed water is an interference, and the method cannot be applied in
these circumstances.  Observations must be made before the point of steam
condensation (detached plume) or after the point of steam dissipation.
Observations must be conducted with the sun angle behind the observer within
a specified angle and with the plume imposed on a contrasting background.
Method 9 is provided in Appendix E for reference.
                                      324

-------
                          REFERENCES FOR SECTION 4
1.   PEDCo Environmental, Inc.  Identification of Parameters That Affect the
     Participate Emissions from Recovery Boilers.  June 1982.

2.   PEDCo Environmental, Inc.  Development of Pilot Inspection System for
     Virginia Air Pollution Control Commission, Interim Results.  November
     1982.

3.   Saunders, G., and B. DeWees.  Observing and Establishing Plant Operating
     Baseline Conditions During Compliance Emission Tests.  PEDCo Environ-
     mental, Inc.

4.   White, H. 0.  Electrostatic Precipitation of Flyash, Part I.  Journal
     of the Air Pollution Control Association.  January 1977.

5.   Hawks, R., and 6. Saunders.  Unpublished data.  PEDCo Environmental, Inc.

6.   U. S. Environmental Protection Agency.  Standards of Performance for
     New Stationary Sources.  EPA-340/1-82-005, June 1982.
                                       325

-------

-------
         APPENDIX A
SUMMARY OF STATE REGULATIONS
                 A-l

-------

-------
  TABLE A-l.  SUMMARY OF SELECTED VISIBLE EMISSION, PARTICULATE, S02, AND TRS EMISSION LIMITS
                        APPLICABLE TO SOURCES WITHIN A KRAFT PULP MILL
State
Alabama






Arizona
Participate
general
process - .
regulations
Class II— exist-
ing and all new
E = 4.10 p°'67
P < 60,000 Ib/h
E = 55.0P0-1-40
P > 60,000 Ib/h
Class I"
exlsting
E = 3.59 P0-62
P £60,000 Ib/h
E = 17.31 P0-16
P > 60,000 Ib/h
E = 4.10P0'67
P <. 60,000 Ib/h
E = 55.0P°-^40
P > 60,000 Ib/h
Visible 2
emission
20% (I)






40% (2)
so2—
general
process







10% of Sentering
the process
S0,--fuel
3
burning
Class I
1.8 lb/106 Btu
Class II
4 lb/106 Btu



New
0.8 Ib S02/106
Btu
Oil or coal
existing
1.0 Ib S02/106
Btu
oil or coal
Particulate--,
fuel burning
Class I—exist-
ing and all new
E = 1.38 Q"°'44
Class II—
existing
E = 3.109 q-°-589
'
Q < 4200 x 106
Btu/h
E = 1.02QU-769
Q > 4200 x 106
Btu/h
E = 17.0Q0'432
Particulate—
kraft pulp
mills4
Rec. furn.
4 Ib/TADP
Smelt tank
0.5 Ib/TADP
lime kiln
0.1 Ib/TADP


Total reduced
sulfur— kraft
pulp mills4
1.2 Ib/TADP5







(continued)

-------
TABLE A-1 (continued)
State
Arkansas



Florida


Georgia


Par t1culate
general
process - ,
regulations
P * 100 Ib/h,
E = 2.5 Ib/h;
P * 1.0 x 106
Ib/h;
E = 100 Ib/h;
P = 1.0 x 106
Ib/h;
E = 700 Ib/h
E = 3.59P0'62
P < 60,000 Ib/h
E = 17.31P0'16
P > 60,000 Ib/h




Visible -
eralsslon
New
20% (1)
Existing
40% (2)


New
20* (1)
Existing
40% (2)
Carbonaceous
fuel
30% (1.5)
New
20% (1)
Existing
40% (2)

so2~
general
process










S0,--fue1
3
burning




> 250 x 106 Btu/h
Liquid fossil
2.75 Ib S02/106
Btu
Solid fossil
6.17 Ib S02/106
Btu
Liquid fossil
< 100 x 106 Btu/h
2.6% S
> 100 x 106 Btu/h
3.0% S

Partlculate--,
fuel burning
Uses general pro-
cess weight rate
for solid fuel

-

Fossil fuel
> 250 x 106 Btu/h
0.1 lb/106 Btu
(2-h avg)
Carbonaceous fuel
> 30 x 106 Btu/h
0.2 lb/106 Btu
Applies to non-
attainment
areas
New
< 10 x 106 Btu/h
0.5 lb/106 Btu/h
10-250 x 106
Btu/h

Partlculate--
kraft pulp
Mills*
Source by source
Rec. furn.
80 - 800 Ib/h
Smelt tank
25 - 55 Ib/h
Lime kiln
40 - 110 Ib/h
3 lb/3000 BLS





Total reduced
sulfur— kraft
pulp mills*




New
•1 ppm or 0.03 Ib/
3000 Ib/BLS
Existing
17.5 ppm or 0.05
lb/3000 Ib BLS
Rec. furn.
Old - 20 ppm
(8% 02)
Hew - 5 ppm
Cross
25 ppm
 (continued)

-------
         TABLE A-1  (continued)

State
Georgia (cont.)














Idaho






Particulate
general
process - ,
regulations















E - 4.10P0'67
P <.60,000 Ib/h
E = 55.0P0-1-40
P > 60,000 Ib/h




Visible -
emission















New
20% (1)
Existing
40% (2)




so2-
general
process















New
20% (1)
Existing
40% (2)




SO,— fuel
3
burning
Solid fossil
c
< 100 x 10° Btu/h

2.5% S
> 100 x 106 Btu/h
3.0% S
Existing
< 10 x 106 Btu/h
0.07 lb/106 Btu/h
10-206o x 106
Btu/h
E . 0.7 (10)0.202
> 2000 x 106
Btu/h
0.24 lb/106 Btu
Residual oil
1.75% S
Distillate oil
Grade 1—0.3% S
Grade 2—0.5% S
Coal
1.0% S

Particulate—,
fuel burning
E • 0.5 (IP-)0'5
> 250 x 106 Btu/h
c
0.10 Ib/lO? Btu











Q 1 10 x 106 .
Btu/h
0.6 lb/106 Btu/h
Q >. 15,000 x 106
Btu/h
0.12 lb/106
Btu/h

Particulate—
kraft pulp
mills4

















'





Total reduced
sulfur— kraft
pulp mills4
Smelt tank

0.0168 Ib TRS/ton
pi c
Di_J
Lime kiln
,40 ppm (10% 02)
















>
         (continued)

-------
TABLE A-1 (continued)

Kentucky












Partlculate
general
process - ,
regulations1
New
E ' 3.59P0'62
P <_ 60,000 Ib/h
E - 1731P0'16
P > 60,000 Ib/h
Existing
E - 55.0P-0-11
40
0.02 gr/ft3
97X efficiency











Visible *
emission
Hew--20Z
40X 2-m1nute
In any 1 h
Existing
40S












sor-
general
process













S0,--fuel
3
	 burning 	
County Class I
Liquid fuel
Y»7.7223X-°-4106
Solid fuel
YM3.8781X"0-4434
County Class II
Liquid fuel
Y»9.4644X"0l374°
Solid fuel
Y=14.1967X'°-3740
County Class III
Liquid fuel
Y'8.060X"0<2436
Solid fuel
Y=12.2539X'°'2432
County Class IV
Liquid fuel
Y«7.3639X"°'1347
Solid fuel
Y=10.8875X'°'1338
County Class V
Liquid fuel
Y-8.0189X"0'1260
articulate— ,
fuel burning
riorlty I
-0.9634Q-0-2356
riorlty II
.1.2825X-0-2330
riorlty III
Y=1.3152X-0'2159












Partlculate—
Scraft pulp
mills*
Rec. furn.
3.5 Ib/ton
Lime kiln
1.0 Ib/ton
Smelt tank
0.5 Ib/ton












Total reduced
sulfur—kraft
pulp mills*
Rec. furn.
15 ppm arlth. avg
40 ppm for more
than 60 tnln. In
any 24 h
Thermal oxidation
98X efficiency












 (continued)

-------
        TABLE A-1 (continued)
State 	 '
Kentucky (cont.)














Particulate
general
process - ,
regulations











-



Visible 2
emission


<












so2-
general
process















S0,,~fuel
3
burning
Solid fuel
Y=12.028«-°-1260
NGW
Liquid
0.8 lb/106 Btu
(2h avg.)
Solid
1.2 lb/106 Btu
(2h avg.)
Existing
Priority I
Same as new
Priority II
Liquid
1.5 lb/106 Btu
Solid
2.0 lb/106 Btu
Priority III
Liquid
2.0 lb/106 Btu
^ Solid
3.5 lb/106 Btu
Particulate—,
fuel burning















Particulate—
kraft pulp
mills*















Total reduced
sulfur—kraft
pulp mills"















>
         (continued)

-------
         TABLE A-l (continued)
State
Louisiana



Ma'ine



Maryland
Partlculate
. general
process - .
regulations
E - 4.10P0'67
P <. 60,000 Ib/h
E - 55.0P0'^40
P > 60,000 Ib/h

E = 3.59P0'62
P <_ 60,000 Ib/h
E = 17.31P0'16
P > 60,000 Ib/h

Hew
0.05 gr/SCFD
Existing
E = 55. OP0' "40
P < 60.000 Ib/h
0.05 gr/SCFD
Visible .
emission
20% (1)



40X (2)



20% (I)
^so2-
general
process
2000 ppm







New
500 ppm
Existing
2000 ppm
S0,~fuel
3
burning
2000 ppm



Area 1 - 6
2.5X S
Area 7
after 11/1/75
1.5% S
after 11/1/85
1.0% S
>. 13 x 106 Btu/h
Residual oil
2.0% S
Distillate oil
Process gas
0.3% S
Partlculate--,
fuel burning
0.6 lb/106 Btu



3-10 x 106 Btu/h
0.6 lb/106 Btu
> 150 x 106
Btu/h
0.3 lb/106 Btu

New
13 i Q <. 25 x
106 Btu
25 < 0 <_ 250 x
106 Btu/h
Partlculate—
kraft pulp
mlllsl
Rec. furn.
4 Ib/TADP
Smelt tank
0.5 Ib/TADP
Lime kiln
1.0 Ib/TADP
Rec. furn.
4 Ib/TADP
Smelt tank
0.5 Ib/TADP
Lime kiln
1.0 Ib/TADP

Total reduced
sulfur—kraft
pulp mills'








0.6 lb/TODP6
>
00
          (continued)

-------
TABLE A-l (continued)
Stdte
Maryland (cont.)

Michigan




Partlculate
general
process - ,
regulations1
P > 60,000 Ib/h

E = 4.10P0'67
P £60,000 Ib/h
E = 55.0P0-1-40
P > 60,000 Ib/h




Visible -
emission


(2)



•-
so2-
general
process







SO,— fuel
3
burning
>_ 13 unit actual
and >_ 100
plant design
capacity
Solid fuel
3.5 Ib S02/
106 Btu
< 500,000 Ib
steam/h
coal 1.5% S,
2.4 Ib S02/106
Btu
oil 1.5* S,
1.7 lb/106 Btu
> 500,000 Ib
steam/h
coal 1.0* S,
1.6 lb/106 Btu
oil 1.0*. S
1.1 lb/106 Btu
Particulate— 3
fuel burning
E = log 10
[6.597538Q"0-3]
> 250 x 106 Btu/h
6.1 x 106 Btu/h
Existing
< 10 x 106 Btu/h
0.6 lb/106 Btu/h
> 10 x 106 Btu/h
E = 1.025985Q"0'23
Pulverized coal
<_ 100 x 103 Ib
steam/h
0.3 lb/1000 Ib gas
Other
'articulate—
kraft pulp
mills'!

297

0-100 0.65 lb/1000 Ib gas
100-300 0.64-0.45 lb/1000 Ib gas








Total reduced
sulfur— kraft
pulp mills'







 (continued)

-------
TABLE A-1 (continued)
State
Minnesota








Partlculate
general
process -
regulations1
E = 4.10P0'67
P <. 60,000 Ib/h
E ' 55.0P0ll*40
P > 60,000 Ib/h








Visible
emission2
New
201 (1)
Existing
60S (3)








so2-
general
process









S02~fuel
burning3
New
< 250 x 106 Btu/h
Liquid
2.0 lb/106 Btu
Solid
4.0 lb/106 Btu
> 250 x 105 Btu/h
Liquid
0.8 lb/106 Btu
Solid
1.2 lb/106 Btu
Existing
< 250 x 106 Btu/h
Liquid
2.0 lb/106 Btu
Solid
4.0 lb/106 Btu
> 250 x 106 Btu/h
Liquid
1.6 lb/106 Btu
Solid
3.0 lb/106 Btu
Partlculate—
fuel burning3
New
0.4 lb/106 Btu
Existing
0.6 lb/106 Btu








Particulate—
kraft pulp
mills*









Total reduced
sulfur—kraft
pulp mills''









(continued)

-------
TABLE A-1 (continued)

State
Mississippi







Montana









Participate
general
process - .
regulations
E=4.10P°'67







E = 4.10P0-67
P £60,000 Ib/h
E = 55. OP0-1-^
P > 60,000 Ib/h







Visible 2
emission
40% (2)







New
20%
Existing
(2)







so2-
general
process
New
500 ppm
Existing
2000 ppm




Concentrations
shall not exceed
2 ppm at any
time





.

SO,— fuel
3
burning
New
Mod. units
2.4 Ib S02/106
Btu
Existing
< avg annual .
emission rate
for 1970 units
1.0 1b/106 Btu










Particulate—,
fuel burning
£ 10 x 106 Btu/h
0.6 lb/106 Btu
> 115 x 106 Btu/h
0.1 lb/106 Btu




New
£ 10 x 106 Btu/h
0.6 lb/106 Btu
100 x 106 Btu/h
0.35 lb/106 Btu
1000 x 106 Btu/h
0.2 lb/106 Btu
>. 10,000 x 106
Btu/h
0.12 lb/106 Btu

Particulate—
kraft pulp
mills*
Rec. furn.
4 Ib/TADP
Smelt tank
0.5 Ib/TADP
Lime kiln
1.0 Ib/TADP













Total reduced
sulfur--kraft
pulp mills*








Rec. furn.
0.087 lb/1000 Ib
BLS or 17.5 ppm







(continued)

-------
 TABLE A-1  (continued)
State
Pennsylvania
(cent.)












Partkulate
general
process -
regulations1













Visible
emission'








.




soz~
genera]
process 	













S02--fue1
burning3 	
> 250 x 106 Btu/h
Inner—O.es
Outer-- 1.2%
Fuel oil --Com.
No. 2 & lighter
Inner— 0.2%
Outer— 0.3%
No. 4,5,6, and
heavier
Inner- -0.5%
Outer— 1. OX
Noncommercial
Inner— 0.62
Outer-1.2%
Partlculate—
fuel burning3













Partlculate—
kraft pulp
mills''













Total reduced
sulfur—kraft













(continued)

-------
         TABLE A-1 (continued)

State
Montana (cont.)









New Hampshire










Particulate
general
process - .
regulations










New
E = 4.10P°'67
P < 60,000 Ib/h
nil
E = 55. OP" -40
P > 60,000 Ib/h
Existing
EV5.05P°'67
P £60,000 Ib/h
E = 66. OP0' ^48
P > 60,000 Ib/h

Visible -
emission










New
20%
Existing
C
< 250 x 10° Btu
40%
>' 250 x 106 Btu
20%




. S02"
general
process






















SO,— fuel
3
burning










Solid
2.8 Ibs S/106
Btu after

4/15/80
1.5 Ibs/ 106 Btu
liquid
#2 - 4% S
#4 - 11 S
#5 4 #6 - 2% S


Particulate—,
fuel burning
Existing
<. 10 x 106 Btu/h
0.6 lb/106 Btu
100 x 106 Btu
0.4 lb/106 Btu
1000 x 106 Btu
0.28 lb/106 Btu
>. 10,000 x 106
Btu
0.19 lb/106 Btu
New
Q < 10 x 106
Btu/h
£
0.6 lb/10° Btu
10 < Q < 250 x
106 Btu/h
E=1.0286Q~°-2341
Q > 250 x 106
Btu/h
0.1 lb/106 Btu

Particulate—
kraft pulp
mills*










Rec. furn.
4 Ib/TADP
Smelt tank

0.5 Ib/TADP
Lime kiln
1.0 Ib/TADP





Total reduced
sulfur— kraft
pulp mills4










2 Ibs/TADP










V-»
CO
        (continued)

-------
TABLE  A-1  (continued)

State
Hew Hampshire
(cent.)








New York




















Participate
general
process - ,
regulations










E = 0.24P0'665
P < 100,000 Ib/h
E«3gp0.08250
P > 100,000 Ib/h


















Visible ,
emission










20% (1)





















so2«
general
process
































S0,,--fuel
<• 3
burning










By area
New York City
Fuel oil
0.2* to'l.OX
Coal— 0.2 to
0.6 lb/106 Btu
Rest of State
New
> 250 x 106
Btu/h
Fuel oil— 0.75X S
Coal
0.6 Ib S/106 Btu
New and Existing
< 250 x 106
Btu/h
Fuel oil -23 S
Coal— 2.5 lb/106
Btu S max or
1.9 lb/106 Btu
S avg.

Particulate--,
fuel burning
Existing
0 <. 10 x 106
Btu/h
0.6 lb/106 Btu
10 < 10,000 x
106 Btu/h
E=0.8803Q-°<1665
Q > 10,000 x 106
Btu/h
0.19 lb/106 Btu
1-10 x 106 Btu/h
0.62 lb/106 Btu
10-10,000 x 106
Btu/h
E = LOO.'0'22
Coal installa-
tions
< 250 x 106 Btu/h
Prior to 8/11/72
E - LOO.'0'22
Coal and oil In-
stallations
> 250 x 106 Btu/h
After 8/11/72
0.1 lb/106 Btu
50-250 x 106 Btu/f
011-0.2 lb/106
Btu




Particulate-
kraft pulp
mills4





























•


Total reduced
sulfur— kraft
pulo wills*































(continued)

-------
       TABLE A-1 (continued)
State 	
North Cdrol Ififl











Ohio











Particulate
general
process - .
regulations
E = 4.10P0'67
P < 60,000 Ib/h
E » 55.0P0<1-40
P^ 60,000 Ib/h








E = 4.10P0'67
P <_ 60,000 Ib/h
E = SS.OP0'1^
P > 60,000 Ib/h








Visible 2
emission 	
New
20% (1)
Existing
40% (2)








20% (1)











so2-
general
process
























S0,--fuel
3
burning
2.3 Ib SOZ/106
Btu

t








Source specific









*

Part1culate--,
fuel burning
Coal, oil
Q < 10 x 106 Btu/h
0.6 lb/106 Btu
10-10,000 Btu/h
E = 1.09Q-0-2594
Q >. 10.000 Btu/h
0.1 lb/106 Btu
Hood
Q < 10 x 106 Btu/h
0.7 lb/106 Btu
Q > 10 x 106 Btu/h
E = 1.1698Q-0-2230
Priority I Region
Q <. 10 x 106 Btu/h
0.6 lb/106 Btu
Q >_ 1000 x 106
Btu/h .
0.1 lb/106 Btu
Priority II & III
Q <_ 10 x 106 Btu/h
0.6 lb/106 Btu
Q >. 1000 x 106
Btu/h
0.15 lb/106 Btu
articulate—
kraft pulp
mills*
Rec. furn.
3.0 Ib/TAOP
Smelt tank
0.6 Ib/TADP
lime kiln
O.S Ib/TADP


















Total reduced
sulfur— kraft
pulp mills'*





'


















m
        (continued)

-------
       TABLE A-1 (continued)
State
Oklahoma























Participate
general
process - .
regulations1
E * 4.10P0'67
P <. 60,000 Ib/h
E = 55.0P0<1i40
P > 60,000 Ib/h




















Visible
emission'
20%























so2-
general
process
Kraft- pulp
18 Ib/TADP






















S02--fue1
burning3
























Particulate—
fuel burning3
< 10 x 106 Btu/h
0.6 lb/106 Btu
100 x 106 Btu/h
0.35 lb/106 Btu
1000 x 106 Btu/h
0.2 lb/106 Btu
" >. 10,000 x 106
Btu/h
0.1 lb/106 Btu/h
Wood
< 10 x 106 Btu/h
0.60 lb/106 Btu
10-1000 x 106
Btu/h
0.5 lb/106 Btu
1000-10,000
0.35 lb/106 Btu
>. 10,000 x 106
Btu/h
0.15 lb/106 Btu/h
Combined wood/
fossil
> 250 x 106 Btu
0.1. lb/106 Btu
Particulate—
kraft pulp
mills4
























Total reduced
sulfur—kraft
pulp mills4
























CTl
       (continued)

-------
         TABLE A-1 (continued)
State
Oregon






Particulate
general
process -
regulations1
P = 100 Ib/h
0.46 Ib/h
P = 10,000 Ib/h
10.0 Ib/h
P < 60,000 Ib/h
E = 55.0P0-1-40





Visible
emission'
New
20% (1)
Existing
40% (2)






so2~
general
process
1000 ppm






S02--fuel
burning-*
Residual oil
1.75% S
Distillate
Grade 1 0.3%
Grade 2 0.5%
Coal
1% S
New
150 < Q <. 250 x
106 Btu/h
Liquid
1.4 lb/106 Btu
Solid
1.6 lb/106 Btu
> 250 x 106 Btu/h
Liquid
0.8 lb/106 Btu
Solid
1.2 lb/106 Btu
Particulate—
fuel burning3
New
<_ 10 x 106 Btu/h
0.27 lb/106 Btu
> 1000 x 106
Btu/h
0.05 lb/106 Btu
Existing
£ 10 x 106 Btu/h
0.55 lb/106 Btu
> 1000 x 106
Btu/h
0.27 lb/106 Btu



Particulate--
kraft pulp
mills'*
Rec. furn.
4 Ib/TADP
Smelt tank
0.5 Ib/TADP
Lime kiln
1 Ib/TADP





Total reduced
sulfur—kraft
pulp mills4
0.5 Ib/TADP
or 17.5 ppm




-

I
M
>>l
        (continued)

-------
        TABLE A-1 (continued)
State
Pennsylvania
















Participate
general
process - ,
regulations
0.04 gr/ft3 or
A * 6000 E
E = 150,000 -
300,000
0.02 gr/ft3
E > 300,000 ft3











Visible 2
emission
202
3 minutes
1n any In
60% any
time












so2-
general
process
500 ppm
















SO.-fuel
3
burning
3.0 lb/106 Btu
< 50
A = 5.1E0'14
50 - 2000
1.8 lb/106
> 2000
Allegheny Co.
1.0 lb/106 Btu
< 50
A = 1.7 E0-14
50 - 2000
.6 lb/106 Btu
> 2000
SE Penn Air Basin
< 250 x 106 Btu/h
Inner— 1.0%
Outer— 1.2%
Participate--,
fuel burning
< 50 x 106 Btu/h
0.4 lb/106 Btu
50-600 x 106
Btu/h
A = 3.6E-0-56
> 600 x 106 Btu/h
0.1 lb/106 Btu










Partlculate—
kraft pulp
mills'*

















Total reduced
sulfur— kraft
pulp wins'1






,










>

(->
00
       (continued)

-------
         TABLE A-1 (continued)














Parti cul ate
general
process - ,
regulations1
E = 4 10P0.67
P £60,000 Ib/h
p = cc tip" -4(1
P > 60,000 Ib/h






NPU
F^3.59P°-62
P <• fin nnn ih/h
E = 17.31P0'16
P-> fin nnn ih/h
Existing
E - 4.10P0'67
Visible 2
emission

2058
Existing
40%






New
20% (1)
Existing
40% (2)
Wood
> 100 x 106 Btu/h
40%
so2-
general
process













S0,~fuel
3
burning
Class I—
<_ 10 x 106 Btu/h
3.5 lb/106 Btu
> 10 x 106 Btu/h
2.3 lb/106 Btu
Class II—
< 1000 x 106
Btu/h
3.5 lb/106 Btu/h
>_ 1000 x 106 Btu
2.3 lb/106 Btu
Class III--
3.5 lb/106 Btu
< 1000 x 106
Btu/h
1.6 - 5 lb/106
Btu
> 1000 x 106
Btu/h
1.2-5 lb/106 Btu
Particulate— ,
fuel burning
Q < 1300 x 106
0.6 lb/106 Btu
> 1300 x 106
Btu/h
E = 57.84Q-0'637
Except existing
prior to 2/11/71
Q < 10 x 106
Btu/h
0.8 lb/106 Btu


Mau
lien
Q <_ 10 x 106
Btu/h
0-6 lb/106 Btu
10-250 x 106
Btu/h
E= 0.6(i|)°'5566
Particulate--
kraft pulp
mills4
Rec. furn.
2.75 Ib/TADP
Smelt tank
1.0 Ib/TADP
Lime kiln
1.0 Ib/TADP





Rec. furn.
3 Ib/TADP
Smelt tank
0.5 Ib/TADP
Lime kiln
1 Ib/TADP
Total reduced
sulfur— kraft
pulp mills4 -
Rec. furn. (8% 00)
old 20 ppm
new 5 ppm
cross 25 ppm
Dig. system,
evap. system, and
stripper system
5 ppm
Lime kiln
20 ppm (10% 02)
Smelt tank
0.0084 g/ kg BLS





l-»
ID
         (continued)

-------
        TABLE A-1  (continued)
State
Tennessee (cont.)










Texas
Particulate
general
process -
regulations1
P < 60,000 Ib/h
E = SS.OP0'1^
P > 60,000 Ib/h










E = 3.12P0"985
P < 40,000 Ib/h
E = 25.4P0'287
P > 40,000 Ib/h
Visible
emission2











New
20%
Existing
30%
so2-
general
process












S02~fuel
burninq3











Liquid
440 ppm S02
Solid
3.0 lb/106 Btu
i
Particulate—
fuel burninq3
>. 250 x 106 Btu/h,
0.1 lb/106 Btu
Hood
New
Q < 25 x 106
Btu/h
0.33 gr/sdcf
Q > 100 x 106
Btu/h
0.20 gr/sdcf
Existing
Q < 50 x 106
Btu/h
0.33 gr/sdcf
Q >. 100 x 106
Btu/h
0.1 gr/sdcf
E = 0.048 (stack
flow rate
acfm)-0'62
Particulate—
kraft pulp
mills"












Total reduced
sulfur—kraft
pulp mills4












ro
o
        (continued)

-------
TABLE A-1 (continued)

State
Virginia


















Washington









Participate
general
process - .
regulations
E = 4.10P0'67
P £60,000 Ib/h
E = 55.0P0-1-40
P > 60,000 Ib/h















New
0.1 gr/sdcf
Existing
0.2 gr/sdcf







Visible -
emission
20%


















•35%









Cf\
so2-
general
process



















500 ppm










S0,~fue1
3
burning
AQCR 1-6
2.64Q
AQCR 7
Liquid or gas
1.06 Q
Solid
1.52Q












1000 ppm










Particulate--,
fuel burning
Q < 10 x 106
Btu/h
0.6 lb/106 Btu
10-10,000 x 106
Btu/h
E=1.096Q-°'2594
> 10,000 x 106
Btu/h
0.1 lb/106 Btu
AQCR 7
< 100 x 106 Btu/h
0.3 lb/106 Btu
100-10,000 x 106
Btu/h
, - on-0.2386
E = 0.9Q
>. 10,000 x 106
Btu/h
0.1 lb/106 Btu
New
0.1 gr/sdcf
'Existing
0.2 gr/sdcf







Partlculate--
kraft pulp
mills4
Rec. furn.
3 Ib/TADP
Smelt tank
0.75 Ib/TADP
L1me kiln
1 Ib/TADP
Slaker tank
0.3 Ib/TADP











Rec. furn.
0.23 g/dcm
(0.10 gr/dscf)
Smelt tank
0.15 g/kg
(0.30 Ib/ton)
of solids fired




Total reduced
sulfur— kraft
pulp mills*
1.2 Ib/TADP
(daily avg./
quarter)














-

L1me kiln
Not exceed 8 ppm
for two consecu-
tive hours.
Nbt exceed 50 ppm
for a daily aver-
age. After
1/1/85 shall not
exceed 20 ppm for
a daily average.
 (continued)

-------
        TABLE A-l  (continued)
State
Washington (cont.)











Wisconsin



Parti cul ate
general
process - .
regulations












E = 3.59P0"62
P < 60,000 Ib/h
E = 17.31P0'16
P > 60,000 Ib/h



Visible 2
emission












Mew
20%
Existing
40%



sor-
general
process
















S0,--fue1
3
burning












> 250 x 106 Btu/h
Liquid
0.8 lb/106 Btu
Solid
1.2 lb/106 Btu



Partlculate— 3
fuel burning












New
<. 250 x 106 Btu/h
0.15 lb/106 Btu
> 250 x 106 Btu/h
0.10 lb/106 Btu
Existing
ASH APS-1
Max 0.6 lb/106
Btu
Partlculate—
kraft pulp
mills'*
Lime kiln
0.30 g/dcm
(0.13 gr/dscf)
All other sources
0.23 g/dscm
(0.10 gr/dscf)
Fugitive emis-
sions—reasonable
precautions.
Note: All mills
are required to
meet RACT




Total reduced
sulfur— kraft
pulp mills4












17.5 ppm
Rec. furn.



DO
ro
        Footnotes
        1.   E, allowable emissions In Ib/h; P, process weight rate 1n tons/h.
        2.   X )• Rlngelmann number.
        3.
        4.
         5.
         6.
E, or Y allowable emissions In Ib/h; Q, or X heat Input 1n 10° Btu/h.
Rec. furn, recovery furnace; evap system, multi-effect evaporator system; dig system, digester system;  TADP, ton air dried  pulp;
BLS, black liquor solid.
ppm expressed as H,S on a  dry gas basis.
TODP, ton oven-dried unbleached pulp.

-------
         APPENDIX B



EPA REFERENCE METHODS 1-5, 17
                B-l

-------

-------
                             Title 40—Protection of Environmirrt
   APPENDIX A—REFERENCE METHODS

  The reference  methods in  this appendix
are referred to in § 60.8 (Performance Tests)
and § 60.11  (Compliance With Standards
and Maintenance Requirements) of 40 CFR
Part  60,  Subpart  A (General  Provisions).
Specific uses of these reference methods are
described in the standards of performance
contained in  the subparts, beginning with
Subpart D.
  Within  each standard  of performance, a
section titled  "Test Methods  and Proce-
dures"  is provided to (1) identify the test
methods applicable to the facility subject to
the respective standard and (2) identify any
special  instructions or conditions to be fol-
lowed when applying a method to the re-
spective facility.  Such instructions (for ex-
ample, establish sampling rates, volumes, or
temperatures) are to be used  either in addi-
tion to, or as a substitute  for procedures in a
reference method.  Similarly, for  sources
subject to  emission monitoring  require-
ments,  specific  instructions  pertaining to
any use of a reference method are provided
in the subpart or in Appendix B.
  Inclusion  of methods in this appendix is
not intended as an  endorsement or denial of
their  applicability  to sources that are  not
subject to standards of  performance. The
methods are potentially applicable to other
sources; however,  applicability  should be
confirmed by careful and appropriate evalu-
ation of the  conditions  prevalent at such
sources.
  The approach followed  in the formulation
of the reference  methods involves specifica-
tions  for  equipment, procedures, and per-
formance. In concept, a performance specifi-
cation approach  would be preferable in all
methods because this  allows the greatest
flexibility to the user. In practice, however.
this approach is impractical  in  most cases
because performance specifications cannot
be  established. Most of the methods de-
scribed herein, therefore,  involve  specific
equipment  specifications and  procedures,
and only  a few  methods in  this appendix
rely on performance criteria.
  Minor changes in the reference methods
should not necessarily affect  the validity of
the results and it is recognized that alterna-
tive and equivalent methods  exist. Section
60.8 provides authority for the Administra-
tor to  specify or  approve (1)  equivalent
methods,  (2)  alternative  methods, and (3)
minor changes in the methodology of the
reference methods. It should  be clearly un-
                   B-3

-------
Chapter I—Environmental Protection Agency
                                 App. A
derstood that unless otherwise identified all
such methods and changes must have prior
approval of the  Administrator. An owner
employing such methods or deviations from
the reference methods  without  obtaining
prior approval does so at the risk of subse-
quent  disapproval and retestlng with ap-
proved methods.
  Within the  reference methods,  certain
specific equipment or procedures are recog-
nized as being acceptable or potentially ac-
ceptable and are specifically  identified  in
the methods. The items identified as accept-
able options may be used without approval
but must be identified in  the test report.
The potentially approvable options are cited
as "subject to the approval of the Adminis-
trator" or as "or equivalent." Such poten-
tially approvable  techniques or alternatives
may be used at the discretion  of the owner
without prior approval.  However, detailed
descriptions for applying these potentially
approvable techniques or  alternatives are
not provided in the reference methods. Also,
the potentially approvable options  are not
necessarily  acceptable in  all • applications.
Therefore, an owner electing to use such po-
tentially approvable techniques or alterna-
tives is responsible for: (1) assuring that the
techniques  or alternatives are  in fact appli-
cable and are properly executed; (2) includ-
ing a written description of the alternative
method in  the  test report  (the  written
method must be clear and  must be  capable
of being performed  without  additional in-
struction,  and  the  the degree  of detail
should be similar to the detail contained in
the reference methods); and (3)  providing
any rationale or supporting data necessary
to show the validity of the alternative in the
particular application. Failure to meet these
requirements can result in the Administra-
tor's disapproval of the alternative.

METHOD 1.—SAMPLE AND VELOCITY TRAVERSES
         POR STATIONARY SOURCES

1. Principle and Applicability
  1.1  Principle. To aid in the representa-
tive measurement  of pollutant  emissions
and/or total volumetric flow rate  from  a
stationary source, a measurement site where
the effluent stream is flowing in a known di-
rection is selected, and the cross-section of
the stack is divided into  a number of equal
areas.  A  traverse  point  is  then  located
within each of these equal areas.
  1.2  Applicability. This method is applica-
ble to  flowing gas streams  in ducts, stacks,
and  flues. The  method  cannot be  used
when:  (1) flow is cyclonic  or  swirling (see
Section 2.4), (2) a stack  is  smaller  than
about  0.30 meter (12  in.)  in  diameter, or
0.071 m2(113 in.2) cross-sectional area, or (3)
the measurement site is less than two stack
or duct diameters downstream or less than a
half diameter upstream from a flow disturb-
ance.
  The requirements of  this  method must be
considered before construction of a  new fa-
cility  from which emissions will be meas-
ured;  failure to do so may require subse-
quent  alterations to the stack or deviation
from the standard procedure.  Cases involv-
ing variants are subject to  approval by the
Administrator,  U.S. Environmental  Protec-
tion Agency.
2. Procedure
  2.1  Selection of Measurement Site. Sam
pling or velocity measurement is performec
at a site located at least eight stack or due'
diameters downstream and two  diameter
upstream from any flow disturbance such a
a  bend, expansion,  or contraction in th>
stack, or from a visible flame. If neeessarj
an alternative location  may be selected, at r
position at least two stack or duct diameter
downstream and  a half diameter upstrear'
from any  flow  disturbance. For a rectangi.
lar cross  section, an  equivalent diamete
(D,) shall be calculated from the followin.
equation,  to determine the upstream an
downstream distances:
                    2LW
                  ~(L+W)
where Z.=length and W=width.
  2.2  Determining the Number of Traver-
Points.
                                       B-4

-------
CO


C71
                          0.5
                        50
                        40
                      O
                      a.

                      LU
                      V)
                      cc
                      LU
                        30
                      oc
                      LU
                      00



                      Z
20
                      z  10
                    1.0
                                                               1.5
                               * FROM POINT OF ANY TYPE OF
                                DISTURBANCE (BEND. EXPANSION. CONTRACTION, ETC.)
                                                          2.0
2.5
\
A
1
T
"f
-B
i
1
1


- -
t

i
'DISTURBANCE

MEASUREMENT
:- SITE


DISTURBANCE
*
TJ

                                DUCT DIAMETERS DOWNSTREAM FROM FLOW DISTURBANCE (DISTANCE B)
                                                                                                      10
                                                                                   m



                                                                                   I

-------
Chapter I—Environmental Protection Agency

  2.2.1  Particulate  Traverses.  When the
eight- and  two-diameter criterion  can be
met, the  minimum number  of  traverse
points shall be: (1) twelve, for circular or
rectangular stacks with diameters (or equiv-
alent diameters) greater than 0.61 meter (24
in.); (2) eight, for circular stacks with  diam-
eters between 0.30  and  0.61 meter (12-24
in.); (3) nine,  for rectangular  stacks with
equivalent diameters between 0.30 and 0.61
meter (12-24 in.).
  When the eight- and two-diameter  crite-
rion cannot be met, the minimum  number
of traverse points is determined from Figure
1-1. Before referring to the figure, however,
determine the distances  from  the chosen
measurement site to the nearest upstream
and downstream  disturbances, and divide
each distance by  the stack diameter or
equivalent diameter, to determine  the dis-
tance in terms of the number of duct diame-
ters. Then, determine  from Figure  1-1 the
minimum number of  traverse  points that
corresponds: (1) to the number of  duct di-
                                App. A
ameters upstream; and (2) to the number of
diameters downstream. Select the higher of
the  two minimum numbers  of  traverse
points, or a greater value, so that for circu-
lar stacks the number is a multiple of 4, and
for rectangular stacks, the number is one of
those shown in Table 1-1.
   TABLE 1-1. CROSS-SECTION LAYOUT FOR
          RECTANGULAR STACKS
Number of traverse points
g '' 	
. 12 	 '. 	
16 	
20 	
25 	
30 	
3Q 	
42 «•
49 	
Matrix layout
3x3
4x3
4x4
5x4
5x5
6x5
6x6
7x6
7x7
                                      B-6

-------
pa
                                    50
      DUCT DIAMETERS UPSTREAM FROM FLOW DISTURBANCE (DISTANCE A)

0.5                 1.0                 1.5                 2.0
                                    40
                                  O
                                  a.
                                  LU
                                  CO
                                  > 30
                                    20
                                  Z
                                  S
                                 Z 10
                                                         T
                                                                                                                   2.5
\
T
A
_t
t
8
A.
—


i
I

DISTURBANCE

MEASUREMENT
- SITE


DISTURBANCE
&
 2         34         5         6         7          8         9

     DUCT DIAMETERS DOWNSTREAM FROM FLOW DISTURBANCE (DISTANCE B)
                                                                                                                   10
                                                                                 f

                                                                                 I

                                        Figure 1-2. Minimum number of traverse points for velocity* (nonparticulate) traverses.

-------
Chapter 1—Environmental Protection Agency
                                                            APR. A
  2.2.2 Velocity   (Non-Particulate)   Tra-
verses. When velocity  or volumetric flow
rate is to be determined (but not participate
matter), the same procedure as that for par-
tlculate traverses (Section 2.2.1) is followed,
except that Figure 1-2 may be used instead
of Figure 1-1.
  2.3  Cross-sectional Layout  and Location
of Traverse Points.-
  2.3.1 Circular Stacks. Locate the traverse
points on two perpendicular  diameters  ac-
cording to Table 1-2 and the example shown
in Figure 1-3. Any equation (for examples,
see  Citations 2 and  3 in  the  Bibliography)
that gives the same values as those in Table
1-2  maybe used in lieu of Table 1-2.
  For participate traverses, one of the diam-
eters  must be in a plane containing the
greatest  expected concentration variation,
e.g., after bends, one diameter shall be in
the plane of the bend. This requirement be-
comes less critical as the distance from the
disturbance increases;  therefore,  other  di-
ameter locations may be used,  subject to ap-
proval of the Administrator.
                            In addition for stacks having diameters
                          greater than 0.61  m (24 in.) no  traverse
                          points shall be  located within 2.5 centi-
                          meters (1.00 in.) of the stack walls; and for
                          stack diameters equal to or less than 0.61 m
                          (24 in.), no traverse points shall be located
                          within 1.3 cm (0.50 in.) of the stack walls.
                          To meet  these criteria, observe  the proce-
                          dures given below.
                            2.3.1.1   Stacks  With Diameters  Greater
                          Than 0.61 m (24  in.). When any  of the tra-
                          verse points as located in Section  2.3.1 fall
                          within 2.5 cm (1.00 in.) of the stack  walls, re-
                          locate them away from the  stack  walls to:
                          (1) a distance of 2.5 cm (1.00 in.); or (2) a
                          distance equal to the nozzle inside diameter,
                          whichever is larger. These   relocated tra-
                          verse points (on each end of a diameter)
                          shall be the "adjusted" traverse points.
                            Whenever  two  successive traverse points
                          are combined to form a single adjusted tra-
                          verse point, treat the adjusted point as two
                          separate traverse points, both in the sam-
                          pling (or  velocity measurement)  procedure,
                          and in recording the data.
   TRAVERSE
     POINT

        1
        2
        3
        4
        S
 DISTANCE.
% of diameter

    4-4
    14.7
    29.5
    70.5
    85.3
    95.6
                Figure 1-3.  Example showing circular stack cross section divided into
                12 equal areas, with location of traverse points indicated.
              TABLE 1-2. LOCATION OF TRAVERSE POINTS IN CIRCULAR STACKS
                      [Percent of stack diameter from inside wall to traverse point]
Traverse point number on a diameter
,
3
6 1™.. 1 	 	 - 	
7 «»»«.«.«..««.« 	
Number of traverse points on a diameter—
2
14.6
85.4
4
6.7
25.0
75.0
83.3
6
4.4
14.6
29.6
70.4
85.4
95.6
8
3.2
10.5
19.4
32.3
67.7
80.6
89.5
10
2.6
8.2
14.6
22.6
34.2
65.8
77.4
12
2.1
6.7
11.8
17.7
25.0
35.6
64.4
14
1.8
5.7
9.9
14.6
20.1
26.9
36.6
16
1.6
4.9
8.5
12.5
16.9
•22.0
28.3
18
1.4
4.4
7.5
10.9
14.6
18.8
23.6
20
1.3
3.9
6.7
9.7
12.9
16.5
20.4
22
1.1
3.5
6.0
8.7
11.6
14.6
18.0
24
1.1
3.2
5.5
7.9
10.5
13.2
16.1
                                         B-8

-------
App. A
   Title 40—Protection of Environment
         TABLE 1-2. LOCATION OF TRAVERSE POINTS IN CIRCULAR STACKS—Continued

                      [Percent of stack diameter from inside wall to traverse point]
Traverse point number on a diameter
8 	 „ 	
g
10 	 :...
11 	
12 	
13
14 	
15
16 	
17 	 ; 	
18 	 	 „ 	
19
20 	 	 	
21
22
23 	 	 	
24

Number of traverse points on a diameter—
2


















4


















6


















8
96.8
















10
85.4
91.8
97.4
	












12
75.0
82.3
88.2
93.3
97.9












14
63.4
73.1
79.9
85.4
90.1
94.3
98.2










16
37.5
62.5
71.7
78.0
83.1
87.5
91.5
95.1
98.4







18
29.6
38.2
61.8
70.4
76.4
81.2
85.4
89.1
92.5
95.6
98.6






20
25.0
30.6
38.8
61.2
69.4
75.0
79.6
83.5
87.1
90.3
93.3
96.1
98.7




22
21.8
26.2
31.5
39.3
60.7
68.5
73.8
78.2
82.0
85.4
88.4
91.3
94.0
96.5
98.9


24
19.4
23.0
27.2
32.3
39.8
60.2
67.7
72.8
77.0
80.6
83.9
86.8
89.5
92.1
94.5
96.8
98.9
  2.3.1.2 Stacks With Diameters Equal  to
or Less Than 0.61 m (24 in.). Follow the pro-
cedure  in Section 2.3.1.1, noting only  that
any "adjusted" points should be relocated
away from the stack walls to: (1) a distance
of 1.3 cm (0.50 in.); or (2) a distance equal to
the nozzle  inside  diameter,  whichever  is
larger.
  2.3.2  Rectangular Stacks. Determine the
number of traverse points as explained  in
Sections 2.1 and 2.2 of this  method. From
Table 1-1, determine the grid configuration.
Divide the stack cross-section into as many
equal rectangular' elemental  areas  as tra-
verse points,  and  then locate a  traverse
point at the centroid of each equal area ac-
cording to the example in Figure 1-4.
  If the tester desires to use more than the
minimum  number   of  traverse   points,
expand the "minimum number of traverse
points" matrix (see Table 1-1) by adding the
extra traverse points along one or the other
or both legs of the matrix; the final matrix
need not be balanced. For example, if a 4x3
"minimum number of points; matrix  were
expanded to  36 points,  the final  matrix
could be 9x4 or 12x3, and would not neces-
sarily have to be 6x6. After constructing the
final  matrix, divide  the stack cross-section
into as many equal rectangular, elemental
areas as traverse points, and locate a tra-
verse point at the centroid  of each equal
area.
  The situation of traverse points being too
close  to the stack walls is not expected  to
arise  with rectangular stacks. If this prob-
lem should ever arise,  the  Administrator
must be contacted  for resolution  of the
matter.
  2.4  Verification of Absence of Cyclonic
Flow. In most stationary sources, the direc-
tion of stack gas flow is essentially parallel
to the stack walls. However, cyclonic flow
may exist (1) after such devices as cyclones
and inertial  demisters  following  venturi
scrubbers, or (2) in stacks having tangential
inlets  or other duct configurations which
tend to induce swirling; in these instances,
the presence or absence of cyclonic flow at
the sampling location must be determined.
The following techniques are acceptable for
this determination.

o

o



0


o
r 	
0

1".

o

/ 1
1
---t-
0 j
I
l~
1 .
0 1
4

o
	 _
0

" ~™~ ™*"

o

Figure 1 -4. Example showing rectangular stack cross
section divided into 12 equal areas, with a traverse
point at centroid of each area.
  Level and zero the manometer. Connect a
Type S pitot tube to the manometer. Posi-
tion the Type S pitot tube at each traverse
point, in succession,  so that the planes of
the face openings of the pitot tube are per-
pendicular  to  the  stack  cross-sectional
plane; when the Type S pitot tube is in this
position, it is at "0° reference." Note the dif-
ferential pressure (Ap) reading at each tra-
verse point. If a null (zero) pitot reading is
obtained at 0° reference at a given traverse
                                         B-9

-------
 Chapter I—Environmental Protection Agency
    \
 point, an acceptable flow condition exists at
 that point. If the pitot reading is not zero at
 0*  reference, rotate the pltot tube (up to
 ±90' yaw angle), until a null reading is ob-
 tained. Carefully determine and record the
 value of the rotation angle (a) to  the near-
 est degree. After the null  technique  has
 been applied at each traverse point, calcu-
 late the average of the absolute values of a;
 assign a values of  0* to those points for
 which no rotation was reauired, and include
 these in the overall  average. If the average
 value of a is greater than 10%  the  overall
 flow condition  in the stack is unacceptable'
 and alternative methodology, subject to the
 approval of the Administrator, must be used
 to perform accurate sample and velocity tra-
 verses.
                                 App. A
 3. Bibliography
  1. Determining Dust  Concentration in  a
 Gas Stream, ASME. Performance Test Code
 No. 27. New York, 1957.
  2. Devorkin, Howard,  et al. Air Pollution
 Source Testing Manual. Air Pollution Con-
 trol District. Los  Angeles,  CA. November
 1S63.
  3. Methods for Determination of Velocity,
 Volume,  Dust and Mist Content of Gases.
 Western  Precipitation Division of Joy Man-
 ufacturing Co. Los Angeles, CA. Bulletin
 WP-50.1968.
  4. Standard Method for Sampling Stacks
 for Partlculate Matter. In: 1971 Book of
 ASTM Standards, Part 23. ASTM Designa-
 tion D-2928-71. Philadelphia, Pa. 1971.
  5. Hanson, H. A., et al.  Particulate Sam-
 pling Strategies for Large  Power Plants In-
 cluding Nonuniform Plow.  USEPA, OBD,
ESRL, Research Triangle  Park, N.C. EPA-
 600/2-76-170, June 1976.
  6. Entropy Environmentalists, Inc. Deter-
mination of the Optimum Number of Sam-
 pling Points: An Analysis of Method 1 Crite-
 ria. Environmental Protection Agency, Re-
 search Triangle  Park, N.C. EPA Contract
 No. 68-01-3172. Task 7.

 METHOD 2—DETERMINATION OP STACK GAS
  VELOCITY AND  VOLUMETRIC  PLOW  RATE
  (TYPE S PITOT TUBE)

 1. Principle and Applicability
  1.1  Principle. The average gas velocity in
 a stack is determined from the gas density
 and from measurement of the average veloc-
 ity head with a Type S (Stausscheibe or re-
 verse type) pitot tube.
  1.2  Applicability. This method  is applica-
 ble for measurement of the average velocity
 of  a gas stream and  for  quantifying  gas
 flow.
  This procedure is not  applicable at mea-
 surement sites which fail to'meet the crite-
 ria of  Method  1,  Section 2.1.  Also,  the
 method cannot be used for direct measure-
 ment in cyclonic or swirling gas streams;
 Section 2.4 of Method 1 shows how to deter-
 mine cyclonic or swirling  flow conditions.
 When unacceptable conditions exist,  alter-
 native procedures, subject  to the approval
 of the Administrator,  U.S. Environmental
 Protection Agency, must  be  employed to
 make accurate flow rate determinations; ex-
 amples of such alternative procedures are:
 <1) to Install straightening  vanes;  (2) to cal-
 culate the total volumetric flow  rate stoi-
 chiometrically, or (3) to move to another
 measurement .site at which the flow is ac-
 ceptable.
 2. Apparatus
  Specifications for the apparatus are given
below. Any  other apparatus that has been
demonstrated  (subject to  approval of the
Administrator) to be capable of meeting the
specifications will be considered acceptable.
                                       B-10

-------
 App. A
   Title 40—Protection of Environment
1.10 • 2.S4 cm*
(9.75-1.0 in.)
            •SUGGESTED (INTERFERENCE FREE)
            PITOT TUBE THERMOCOUPLE SPACING
                       Figure 2-1. Type S pilot tube manometer assembly.
  2.1 Type S PItot Tube. The Type S pitot
tube (Figure 2-1) shall be made of  metal
tubing (e.g. stainless steel). It is recommend-
ed that the external tubing diameter  (di-
mension  D,  Figure  2-2b)  be between 0.48
and 0.95 centimeters (%« and % inch).  There
shall be an equal distance from the base of
each leg of the pitot tube to its face-opening
plane (dimensions PA and PB Figure 2-2b); it
is recommended that this distance be be-
tween  1.05  and  1.50  times  the external
tubing diameter. The face openings of the
pitot  tube shall, preferably, be aligned as
shown in Figure 2-2; however, slight misa-
lignments of the openings are permissible
(see Figure 2-3).
  The Type S pitot tube shall have a known
coefficient, determined as outlined in .Sec-
tion 4. an identification number shall be as-
signed to the pitot tube; this number shall
be permanently marked or engraved on the
body of the tube.
                                      B-ll

-------
r
                                 Chapter I—Environmental Protection Agency
                                              TRANSVERSE
                                               TUBE AXIS
         App. A
                                                            \
                                                                             FACE
                                                                      -•— OPENING-*-
                                                                            PLANES
                                                                         A-SIDE PLANE
LONGITUDINAL
TUBE AXIS *"
t
\
Dt
t
A
B
                                                                                        -     I
                                                                                         PB     I
NOTE:

1.05 Dt*£P< 1.50 Dt
                                                                        B-SIDE PLANE

                                                                          (b)
                                                                  A ORB
                                                                      (c)
                                Figure 2-2. Properly constructed Type S pitot tube, shown in: (a) end view; face opening planes perpendicular
                                to transverse axis; (b) top view; face opening planes parallel to longitudinal axis; (c) side view; both legs of
                                equal length and centerlines coincident, when viewed from both sides. Baseline coefficient values of 0.84 may
                                be assigned to pitot tubes constructed this way.
                                                                        B-12

-------
APP.A
Title 40—Protection of Environment
TRANSVERSE
 TUBE AXIS
LONGITUDINAL
 TUBE AXIS	
                B        FLOW
                      (c)
                                                   B        FLOW
                                                         (d)
                                (e)
                                    B-13

-------
  Chapter I—Environmental Protection Agency
                                  App. A
                                         (f)
                                         (g)
 Figure 2-3. Types of face-opening misalignment that can result from field use or improper construction of
 Type S pltot tubes. These will not affect the baseline value of Cp(s) so long as al and o2 10°, /31 and 02
 S; z 0.32 cm (1/8 In.) and w 0.08 cm (1/32 in.) (citation 11 in Section 6).
   A standard pitot tube may be used instead
 of a Type S, provided  that it meets the
 specifications of Sections 2.7 and  4.2; note,
 however, that the static and impact pres-
 sure holes of standard pitot tubes are. sus-
 ceptible to plugging in particulate-laden gas
 streams. Therefore, whenever a  standard
 pitot tube is used to perform a traverse, ade-
 quate  proof  must be furnished  that the
 openings of the pitot tube have not plugged
 up during the traverse period; this can be
 done by taking a velocity head (Ap) reading
 at the  final traverse point, cleaning out the
 impact and  static holes of the  standard
 pitot tube by "back-purging" with pressur-
 ized air, and then taking another Ap read-
 ing. If the Ap readings made before and
 after the air purge are the same (±5 per-
 cent), the traverse is acceptable. Otherwise,
 reject the run. Note that if Ap at the final
 traverse point is  unsuitably  low,  another
 point may be selected. If "back-purging" at
 regular intervals is part of the procedure,
 then, comparative  Ap readings  shall  be
 taken,  as above, for the last two back purges
' at which suitably high Ap readings are ob-
 served.
   2.2  Differential  Pressure Gauge.  An in-
 clined  manometer  or  equivalent  device  is
 used.  Most sampling  trains .are  equipped
 with a 10-in. (water column) Inclined-verti-
 cal manometer, having 0.01-in. H,O divisions
 on the 0-to 1-in. inclined scale, and  0.1-in.
 H,O  divisions on  the  1- to  10-in. vertical
 scale. This type of manometer  (or  other
 gauge  of equivalent sensitivity) is satisfac-
 tory for the  measurement of Ap values as
low as 1.3 mm (0.05 in.) HaO. However, a dif-
ferential pressure gauge of  greater sensitiv-
ity shall be used (subject to the approval of
the Administrator), if any of the following
is found to be true: (1) the  arithmetic aver-
age of all Ap readings at the traverse points
in the stack is less than 1.3 mm (0.05 in.)
HaO; (2) for traverses of 12 or more points,
more than 10 percent of the individual Ap
readings are below 1.3 mm (0.05 in.) HaO; (3)
for traverses of fewer than 12 points, more
than one  Ap reading is below 1.3 mm (0.05
in.) HjO. Citation 18 in  Section 6 describes
commercially available instrumentation for
the measurement of low-range gas  veloci-
ties.
  As an alternative to criteria (1) through
(3) above, the following  calculation may be
performed to determine the  necessity of
using a more sensitive differential pressure
gauge:

where:
Ap<=Individual velocity head reading at a
   traverse point, mm HaO (in. HaO).
n=Tptal number of traverse points.
                                         B-14

-------
 App. A
   Title 40—Protection of Environment
K=0.13 mm H2O when metrip units are used
    and 0.005 in H,O when English units are
    used.

If T is greater than 1.05, the velocity head
data are unacceptable and a more sensitive
differential pressure gauge must be used.
  NOTE: If differential pressure gauges other
than inclined  manometers are  used (e.g.,
magnehelic gauges), their calibration must
be checked after'each test series. To check
the  calibration of a  differential  pressure
gauge, compare Ap readings  of the gauge
with those of  a gauge-oil manometer at a
minimum  of  three points,  approximately
representing the range of Ap values in the
stack.  If,  at each point, the values of Ap as
read by the differential pressure gauge and
gauge-oil'manometer agree to within 5 per-
cent, the differential  pressure gauge shall
be considered  to be in proper calibration.
Otherwise, the test series shall either be
voided, or procedures  to adjust the meas-
ured Ap values and final results  shall be
used subject to the approval of the Adminis-
trator.
  2.3  Temperature  Gauge. A  thermocou-
ple, liquid-filled bulb thermometer, bimetal-
lic thermometer, mercury-in-glass thermom-
eter, or other gauge, capable of measuring
temperature  to within 1.5 percent of the
minimum absolute stack temperature shall
be used. The temperature gauge shall be at-
tached to the  pitot  tube such  that the
sensor tip does not touch any metal; the
gauge  shall be in  an  interference-free  ar-
rangement with respect to the pitot tube
face openings  (see Figure  2-1  and also
Figure 2-7 in Section 4). Alternate positions
may be used if the pitot tube-temperature
gauge system is calibrated according to the
procedure of Section 4. Provided that a dif-
ference of not  more than  1 percent in the
average velocity measurement is introduced,
the temperature gauge need not be attached
to the  pitot tube; this  alternative is subject
to the approval of the Administrator.
  2.4  Pressure Probe and Gauge. A piezo-
meter tube and mercury-  or water-filled U-
tube manometer capable of measuring stack
pressure to within 2.5 mm (0.1  in.) Hg is
used. The static tap of a standard type pitot
tube or one leg of a Type S pitot tube with
the face opening planes positioned parallel
to the gas flow may also be used as the pres-
sure probe.
  2.5  Barometer.  A mercury, aneroid,  or
other barometer capable  of measuring at-
mospheric pressure to within 2.5  mm Hg
(0.1 in. Hg) may be used. In many cases, the
barometric reading may be obtained from a
nearby national weather service station, in
which  case the station value (which is the
absolute barometric pressure) shall be re-
quested and an adjustment for elevation dif-
ferences between the weather station and
the sampling point shall be applied at a rate
of minus 2.5 mm (0.1 in.) Hg per 30-meter
(100 foot) elevation  increase or vice-versa
for elevation decrease.
  2.6 Gas Density  Determination  Equip-
ment. Method 3 equipment, if needed (see
Section 3.6), to determine the stack gas dry
molecular weight, and Reference Method 4
or Method 5  equipment  for  moisture con-
tent determination; other methods may be
used subject to approval of the Administra-
tor.
  2.7 Calibration Pilot Tube. When calibra-
tion of the Type S pitot tube is necessary
(see Section 4), a standard pitot tube is used
as  a reference;  The  standard  pitot tube
shall, preferably, have a known coefficient,
obtained either (1) directly from the Nation-
al Bureau of Standards, Route 270,  Quince
Orchard Road, Gaithersburg, Maryland, or
(2) by calibration against another standard
pitot tube with an  NBS-traceable  coeffi-
cient. Alternatively, a standard pitot tube
designed according to the criteria given in
2.7.1 through  2.7.5 below and illustrated in
Figure 2-4 (see also Citations 7, 8, and 17 in
Section 6) may be used. Pitot tubes designed
according to these specifications will have
baseline coefficients of about 0.99±0.01.
  2.7.1  Hemispherical (shown in Figure 2-
4), ellipsoidal,  or conical tip.
  2.7.2  A  minimum   of  six   diameters
straight run (based upon D, the external di-
ameter of the  tube) between the tip and the
static pressure holes.
  2.7.3  A  minimum  of eight  diameters
straight run  between  the  static pressure
holes and the centerline of  the external
tube, following the 90 degree bend.
  2.7.4  Static pressure holes  of equal size
(approximately 0.1 Z»,  equally spaced in a
piezometer ring configuration.
  2.7.5  Ninety degree bend, with curved or
mitered junction.
  2.8  Differential Pressure  Gauge for Type
S Pitot Tube Calibration. An inclined mano-
meter or equivalent is used. If the single-ve-
locity calibration technique is employed (see
Section 4.1.2.3), the calibration differential
pressure gauge  shall  be readable  to the
nearest 0.13 mm H,O (0.005  in. H,O). For
multivelocity  calibrations, the gauge shall
be readable to  the nearest 0.13 mm H,O
(0.005 in HaO) for Ap values between 1.3 and
25 mm  HaO (0.05 and 1.0 in.  H,O),  and to
the nearest 1.3 mm H,O  (0.05 in.  H3O) for
Ap values above 25 mm H»O (1.0 in. H,O). A
special,  more  sensitive gauge will  be re-
quired to read Ap values below 1.3 mm H,O
10.05 in. H,O]  (see Citation 18 in Section 6).
                                        B-15

-------
00
I

                                          t
                                                       CURVED OR
                                                    MITEREO JUNCTION
STATIC
 HOLES
(-0.10)
                                                        HEMISPHERICAL
                                                             TIP
                       Figure 2-4.  Standard pitot tube design specifications.

-------
App. A

3. Procedure
  3.1  Set up  the apparatus as shown In
Figure 2-1. Capillary tubing or surge tanks
installed  between the manometer and pitot
tube may be  used to dampen  Ap fluctu-
ations. It is recommended, but not required,
that a pretest leak-check be conducted, as
follows: (1) blow through the pitot impact
opening until at least 7.6 cm (3 in.) H.O ve-
locity pressure registers on the manometer;
then, close  off  the impact  opening.  The
pressure  shall remain stable for  at least 15
seconds;  (2) do the same for the  static pres-
sure side, except using suction to obtain the
minimum of 7.6  cm (3 in.) H,O. Other leak-
check procedures, subject to the approval of
the Administrator may be used.
  3.2 Level and zero the  manometer. Be-
cause the manometer level and zero  may
   Title 40—Protection of Environment

drift  due to  vibrations  and  temperature
changes, make periodic checks during the
traverse. Record all necessary data as shown
in the example data sheet (Figure 2-5).
  3.3  Measure the velocity  head and tem-
perature at the traverse points specified by
Method 1. Ensure that the proper differen-
tial pressure gauge  is being used  for the
range of Ap values encountered (see Section
2.2). If it is necessary to change to a more
sensitive gauge, do so, and remeasure the Ap
and temperature readings at each traverse
point. Conduct a post-test leak-check (man-
datory), as described in Section 3.1 above, to
validate the traverse run.
  3.4  Measure the  static pressure in the
stack. One reading is usually adequate.
  3.5  Determine the atmospheric pressure.
                                          B-17

-------
 Chapter I—Environmental Protection Agency
                                         App. A
PLANT.
DATE.
.RUN NO.
STACK DIAMETER OR DIMENSIONS, m(in.)
BAROMETRIC PRESSURE, mm Hg (in. Hg)	
CROSS SECTIONAL AREA. m2(ft2)	
OPERATORS 	
PITOTTUBEI.D.NO.
  AVG. COEFFICIENT. Cp = .
  LAST DATE CALIBRATED.
                             SCHEMATIC OF STACK
                               CROSS SECTION
Traverse
Pi. No.


















Vel. Hd.,Ap
mm (in.) H20


















Stack Temperature
ts,°C<°FI


















Averajt
T$, °K (°B)









,









P9
mm Hg (in.Hg)



















Vf Ap



















                    Figure 2-5. Velocity traverse data.
                                 B-18

-------
App. A
    Title 40—Protection of Environment
  3.6 Determine the stack gas dry molecular
weight. For combustion processes or proc-
esses that emit essentially CO,, O,, CO, and
Na> use Method 3. For processes emitting es-
sentially air, an analysis need not be con-
ducted; use a dry molecular weight of 29.0.
For other processes, other methods, subject
to the approval of the Administrator, must
be used.
  3.7 Obtain  the  moisture content from
Reference Method 4 (or equivalent) or from
Method 5.
  3.8 Determine the  cross-sectional area of
the stack or duct at the sampling location.
Whenever possible, physically measure the
stack  dimensions rather than using blue-
prints.

4. Calibration
  4.1 Type S Pitot Tube. Before its initial
use, carefully examine  the Type  S pitot
tube in top, side,  and end views to verify
that the face  openings  of the tube are
aligned within the specifications illustrated
in Figure 2-2 or 2-3. The pitot tube shall
not be used if it fails  to meet these align-
ment specifications.
  After verifying the  face  opening  align-
ment, measure and record the following di-
mensions of the pitot tube: (a) the external
tubing diameter (dimension D,, Figure 2-2b);
and (b) the base-to-opening plane distances
(dimensions  PA and PB, Figure 2-2b). If D, is
between 0.48 and 0.95 cm (¥ie and % in.) and
if PA and Pe  are equal and between 1.05 and
1.50 Di, there are two possible options: (1)
the pitot tube may be calibrated according
to the procedure outlined in Sections 4.1.2
through 4.1.5 below, or (2) a baseline (isolat-
ed tube) coefficient value of 0.84 may be as-
signed to the pitot tube. Note, however, that
if the pitot tube is part of an assembly, cali-
bration may still be required, despite knowl-
edge of the baseline coefficient value (see
Section 4.1.1).
   If D,, PA, and Pa are outside the specified
 limits, the pitot tube must be calibrated as
 outlined in 4.1.2 through 4.1.5 below.
   4.1.1 Type  S  Pitot  Tube  Assemblies.
 During sample and  velocity traverses,  the
 isolated Type S  pitot tube is not always
 used;  in many  instances,  the pitot tube is
 used in combination  with  other source-sam-
 pling  components (thermocouple,  sampling
 probe, nozzle) as part of an "assembly." The
 presence of other sampling components can
 sometimes affect the baseline value of the
 Type  S pitot tube coefficient (Citation 9 in
 Section 6); therefore an assigned (or other-
 wise known)  baseline coefficient value may
 or may not be valid for a given assembly.
 The  baseline  and   assembly  coefficient
 values will be identical only when the rela-
 tive placement of the components in the as-
 sembly is such that aerodynamic interfer-
 ence  effects are  eliminated. Figures  2-6
 through   2-8  illustrate  interference-free
 component arrangements  for  Type S pitot
 tubes having external tubing diameters be-
 tween 0.48 and 0.95 cm (%e and % in.). Type
 S pitot tube assemblies that fail to meet any
 or all of  the specifications of Figures 2-6
 through 2-8 shall be calibrated according to
 the procedure  outlined  in  Sections  4.1.2
 through 4.1.5 below, and prior  to calibra-
 tion, the values of the intercomponent spac-
 ings (pitot-nozzle, pitot-thermocouple, pitot-
 probe sheath) shall be measured and record-
 ed.
   NOTE: Do not use  any Type S pitot tube
 assembly which is constructed such that the
 impact pressure opening plane of  the pitot
 tube is below the entry plane of the nozzle
 (see Figure 2-6b).
   4.1.2 Calibration  Setup. If the  Type S
 pitot tube is to be calibrated, one leg of the
 tube shall be permanently  marked A,  and
 the other, B. Calibration shall be done in a
• flow system  having  the following essential
 design features:
                                       B-19

-------
Chapter I—Environmental Protection Agency
                                                          APP.A
                  1*.
                 TYPE S PITOT TUBE
                          t x> 1.90 cm (3/4 in.) FOR Dn-1.3 cm (1/2 in.)
                 SAMPLING NOZZLE
            A. IOTTOM VIEW; SHOWING MINIMUM PITOT-NOZZLE SEPARATION.
  SAMPLING
   PROIE
                 SAMPLING
                  NOZZLE
       •"
              JJL
STATIC PRESSURE
 OPENING PLANE
                                                              IMPACT PRESSURE
                                                               OPENING PLANE
              '  TYPES
               PITOT TUBE
                   NOZZLE ENTRY
                      PLANE


•.  SIDE VIEW: TO PREVENT PITOT TUBE
   FROM INTERFERING WITH GAS FLOW
   STREAMLINES APPROACHING THE
   NOZZLE. THE IMPACT PRESSURE
   OPENING PLANE OF THE PITOT TUBE
   SHALL BE EVEN WITH OR ABOVE THE
   NOZZLE ENTRY PLANE.
                                     R-20

-------
                                             THERMOCOUPLE
                                                    -0-
                                               TYPE SPITOT TUBE
                                   SAMPLE PROBE
THERMOCOUPLE
Z> S.Mem
Win.) "
i
rc 	 	 - 	 	 c 	 ^

                                                                                                                         •?
                                                                                                                         >
                                                                                             TYPE SPITOT TUBE
                              , SAMPLE PROBE
                                                             CD
CO
I
ro
Figure 2-7. Proper thermocouple placement to prevent interference;
Dr between 0.48 and 0.95 cm (3/16 and 3/8 in.).
                                                                             TYPE SPITOT TUBE
                                                                     ilii I
                                                              SAMPLE PROBE
                                Y >7.62 cm (3 in.)

                                                                                                                         3
                                Figure 2-8.  Minimum pitot-sample probe separation needed to prevent interference;   |
                                n+ hotu/ppn n 48 anri 0 95 rm (3/16 and 3/8 in.).                                       ~

-------
 Chapter I—environmental Protection Agency
                                  App. A
   4.1.2.1  The flowing gas stream must be
 confined to a duct of definite cross-sectional
 area, either circular or rectangular. For cir-
 cular cross-sections, the minimum duct di-
 ameter shall be 30.5 cm (12 in.); for rectan-
 gular cross-sections, the width (shorter side).
 shall be at least 25.4 cm (10 in.).
   4.1.2.2  The cross-sectional  area of  the
 calibration duct must be constant over a dis-
 tance of 10 or more  duct diameters. For a
 rectangular cross-section, use an equivalent
 diameter,  calculated  from the  following
 equation, to determine the number of duct
 diameters:
               ».=
                    2 7,ir
                            Kquaticm 2-1
 where:
 A=Equivalent diameter
 L=Length
 W= Width
  To ensure the presence of stable, fully de-
 veloped flow patterns at the calibration site,
 or "test section," the site must be located at
 least eight diameters downstream and two
 diameters upstream from the nearest distur-
 bances.
  NOTE: The eight- and two-diameter crite-
 ria are not absolute; other test section loca-
 tions may be used (subject to approval  of
 the Administrator), provided that  the flow
 at the test site is stable  and  demonstrably
 parallel to the duct axis.
  4.1.2.3  The flow system shall have the ca-
 pacity to generate a test-section velocity
 around 915 m/min (3,000 ft/min). This ve-
 locity must be constant with time to guaran-
 tee steady flow during calibration. Note that
 Type S pitot tube coefficients obtained by
 single-velocity  calibration at  915  m/min
 (3,000  ft/min) will generally  be  valid  to
 within ±3 percent for the measurement  of
 velocities above 305 m/min (1,000 ft/min)
 and to within ±5 to 6 percent for the mea-
 surement of velocities between 180 and 305
 m/min (600  and 1,000  ft/min).  If a more
 precise correlation between Cf and velocity
 Is desired, the flow system shall have the ca-
pacity to generate  at  least four  distinct,
time-invariant test-section velocities cover-
ing the velocity range from 180 to 1,525 m/
min  (600 to  5,000 ft/min), and calibration
data shall be taken at regular velocity inter-
vals over this range (see Citations 9 and 14
in Section 6 for details).
  4.1.2.4  Two entry ports, one each for the
standard and Type S pitot tubes,  shall be
cut in the test section; the standard pitot
entry port shall be located slightly down-
stream of  the Type S  port, so  that  the
 standard and Type S impact openings will
 lie in the same cross-sectional plane during
 .calibration. To facilitate alignment of the
 pitot tubes during calibration, it is advisable
 that the test section be constructed of plex-
 iglas or some other transparent material.
  4.1.3  Calibration Procedure.  Note that
 this procedure is a general one and must not
 be used  without first referring to the special
 considerations presented in  Section 4.1.5.
 Note also that this procedure applies only
 to single-velocity calibration. To obtain cali-
 bration  data for the A and B sides of the
 Type S pitot tube, proceed as follows:
  4.1.3.1   Make sure that the manometer is
 properly filled and that the oil is free from
 contamination and is of the proper  density.
 Inspect and leak-check all pitot lines; repair
 or replace if necessary.
  4.1.3.2   Level and zero the  manometer.
 Turn on the fan and allow the flow to stabi-
 lize. Seal the Type S entry port.
  4.1.3.3   Ensure that  the  manometer  is
 level and zeroed. Position the standard pitot
 tube at the calibration point (determined as
 outlined in Section 4.1.5.1), and  align the
 tube'so  that its tip is pointed directly into
 the flow. Particular car should be taken in
 aligning  the tube to avoid yaw and pitch
 angles. Make sure that the entry port sur-
 rounding the tube is properly sealed.
  4.1.3.4   Read A",,,, and record its value in a
 data  table  similar  to  the one shown in
 Figure 2-9. Remove the standard pitot tube
 from the duct and  disconnect it  from the
 manometer. Seal the standard entry  port.
  4.1.3.5  Connect the Type S pitot  tube to
 the  manometer.  Open the Type S entry
 port. Check the manometer level and zero.
 Insert and align the Type  S  pitot  tube so
 that its A side impact opening is at the same
 point as  was the standard pitot tube and  is
 pointed  directly  into the flow. Make sure
 that the  entry  port surrounding the tube  is
 properly sealed.
  4.1.3.6  Read Ap. and enter its value in the
 data table. Remove the Type S pitot tube
 from the duct and  disconnect it  from the
 manometer.
  4.1.3.7  Repeat  steps  4.1.3.3  through
 4.1.3.6 above until three pairs of Ap readings
 have been obtained.
  4.1.3.8  Repeat  steps  4.1.3.3  through
 4.1.3.7 above for  the B side of the Type  S
 pitot tube.
  4.1.3.9  Perform calculations, as described
 in Section 4.1.4 below.
  4.1.4 Calculations.
  4.1.4.1  For each  of the six pairs of Ap
readings  (i.e., three  from  side A and three
from side B) obtained in Section 4.1.3 above,
calculate the value of the  Type S pitot tube
coefficient as follows:
                                        B-22

-------
App. A
  PITOTTUBE IDENTIFICATION NUMBER:
  CALIBRATED BY:,	:	
 Title 40—Protection of Environment
	DATE:	

RUN NO.
1
2
3
"A" SIDE CALIBRATION
Apstd
cm H20
(in. HaO)




Apfe)
cmH20

-------
 Chapter I—Environmental Protection Agency
                                  App. A
                         'Ap.,.1

                            P*

                            Equation 2-2
 where:
 C,d)=Type S pitot tube coefficient
 C,(,U)=Standard pitot tube coefficient; use
    0.99 if the coefficient is unknown and
    the tube is designed according to the cri-
    teria of Sections 2.7.1  to  2.7.5 of this
    method.
 Ap.u=Velocity head measured by the stand-
    ard pitot tube, cm H,O (in. H,O)
 Ap.=Velocity head measured by the Type S
    pitot tube, cm H,O (in H,O)

  4.1.4.2  Calculate Cp (side A), the mean A-
 side coefficient, and C? (side B), the mean B-
 slde coefficient:  calculate the difference be-
 tween these two average values.
  4.1.4.3  Calculate the deviation of each of
 the three A-side values of CPo) fronvQ, (side
 A), and the deviation of each B-side value of
 Cft,) from Cf Cf  (side B). Use the following
 equation:
      Deviation=Cpl.)—CP(A or B)

                            Equation 2-3
  4.1.4.4  Calculate S. the average deviation
from the mean, for both the A and B sides
of the pitot tube. Use the following equa-
tion:
 a (side A or B) =—
                            Equation 2-4
  4.1.4.5  Use the Type S pitot tube only if
the values of S (side A) and S (side B) are
less than or equal to 0.01 and if the absolute
value of the difference between CP (A) and
Cf (B) is 0.01 or less.
  4.1.5 Special considerations.
  4.1.5.1  Selection of calibration point.
  4.1.5.1.1  When an isolated Type S pitot
tube is calibrated, select a calibration point
at or near the center of the duct, and follow
the procedures  outlined in Sections 4.1.3
and 4.1.4  above. The Type S  pitot  coeffi-
cients so obtained, i.e., Cp (side A) and CP
(side B), will be valid, so long as either: (1)
the  isolated pitot tube  is used; or (2) the
pitot tube is  used with  other  components
(nozzle, thermocouple, sample probe) in an
arrangement that is free from aerodynamic
interference  effects   (see   Figures  2-6
through 2-8).
  4.1.5.1.2  For Type  S  pitot tube-thermo-
couple   combinations   (without   sample
probe), select a calibration point at or near
the center of the duct, and follow the proce-
dures  outlined in Sections 4.1.3  and 4.1.4
above. The  coefficients so obtained will be
valid so long as the pitot tube-thermocouple
combination is used by itself  or with other
components in an interference-free arrange-
ment (Figures 2-6, and 2-8).
  4.1.5.1.3  For   assemblies   with  sample
probes, the calibration point  should be lo-
cated at or near the center of the duct; how-
ever, insertion of  a probe sheath  into a
small duct may cause significant cross-sec-
tional  area blockage and yield incorrect co-
efficient values (Citation 9  in  Section 6).
Therefore, to minimize the blockage effect,
the  calibration point may be a few Inches
off-center if necessary. The actual blockage
.effect  will be negligible  when the theoreti-
cal blockage, as determined by  a projected-
area model of the probe sheath, is 2 percent
or less of the duct cross-sectional area for
assemblies without external sheaths (Figure
2-10a), and  3 percent  or less for assemblies
with external sheaths (Figure  2-10b).
  4.1.5.2 For  those  probe  assemblies in
which  pitot tube-nozzle  interference  is a
factor  (i.e.,  those in which the  pitot-nozzle
separation distance fails to meet the specifi-
cation illustrated in Figure 2-6a), the value
of C?M depends upon the amount of free-
space  between the  tube and  nozzle, and
therefore is a function  of nozzle size. In
these instances, separate calibrations shall
be performed with each of the commonly
used nozzle sizes in place. Note that the
single-velocity calibration technique is ac-
ceptable for this purpose, even  though the
larger  nozzle sizes (>0.635 cm or 'A in.) are
not  ordinarily used for isokinetic sampling
at velocities around 915 m/min (3,000 ft/
min), which is the calibration velocity; note
also  that it  is not necessary to draw an Iso-
kinetic sample during calibration  (see Cita-
tion  19 in Section 6).
  4.1.5.3  For a probe  assembly  constructed
such that its pitot tube is always used in the
same orientation, only one side  of the pitot
tube need be calibrated (the side which will
face  the flow).  The pitot tube must still
meet the alignment specifications of Figure
2-2 or 2-3, however, and must have, an aver-
age  deviation  (6) value of 0.01  or less (see
Section 4.1.4.4).
                                        B-24

-------
03
I
ro
en
                                           (a)
                                                      ESTIMATED

                                                      SHEATH

                                                      BLOCKAGE
  ElxW   "I

lUCT AREAj
                        (b)
x 100
                                    Figure 2-10.  Projected-area m.odels for typical pitot tube assemblies.
a
5*

8.

f

f

•

-------
 Chapter I—Environmental Protection Agency
                                 App. A
 Figure 2-10. Projected-area models for typi-
 cal pitot tube assemblies.
  4.1.6  Field Use and Recalibration.
  4.1.6.1  Field Use.
  4.1.6.1.1  When a Type S pitot tube (iso-
 lated tube or assembly) is used in the field,
 the appropriate  coefficient value (whether
 assigned or obtained by calibration) shall be
 used to  perform velocity calculations.  For
 calibrated Type S pitot tubes, the A side co-
 efficient shall be used when  the A side of
 the tube faces the flow, and the B side coef-
 ficient shall be used  when the B side faces
 the flow; alternatively, the arithmetic aver-
 age of the  A and B  side coefficient  values
 may be used, irrespective of which side faces
 the flow.
  4.1.6.1.2  When a probe assembly is used
 to sample a small duct (12 to 36 in. in diame-
 ter), the probe sheath sometimes blocks a
 significant  part  of the  duct cross-section,
 causing a reduction in the effective value of
 Cr(j}. Consult Citation 9 in Section 6 for de-
 tails. Conventional pitot-sampling probe as-
 semblies are not recommended  for use in
 ducts having inside diameters smaller than
 12 Inches (Citation  16 in Section 6).
  4.1.6.2  Recalibration.
  4.1.6.2.1  Isolated Pitot Tubes. After each
 field use, the pitot tube shall be carefully
 reexamined in  top, side, and  end views. If
 the  pitot face openings are still aligned
 within  the specifications  illustrated  in
 Figure 2-2 or 2-3, it can be assumed that the
 baseline coefficient of the pitot tube has not
 changed.  If, however, the  tube  has been
 damanged to the' extent that it no longer
 meets the specifications of Figure 2-2 or 2-
 3, the damage shall either be repaired to re-
 store proper alignment of the  face openings
 or the tube shall be discarded.
  4.1.6.2.2 Pitot  Tube Assemblies.  After
 each field use, check the face opening align-
 ment of  the  pitot  tube,  as in Section
 4.1.6.2.1;  also, remeasure  the  intercompon-
 ent spacings of the assembly. If the inter-
 component spaclngs have not changed and
 the face opening  alignment is acceptable, it
 can be assumed that  the  coefficient of the
 assembly has not changed. If the face open-
 Ing alignment is no longer within the speci-
 fications of  Figures 2-2 or 2-3, either  repair
 the damage or  replace the pitot  tube (cali-
brating the  new assembly, If necessary). If
 the intercomponent spacings have changed,
restore the  original spacings or recalibrate
 the assembly.
  4.2  Standard pitot  tube (if applicable). If
 a standard pitot tube Is used for the velocity
 traverse, the tube shall be  constructed ac-
cording to the 'criteria of Section 2.7 and
shall be assigned a baseline coefficient value
of 0.99. If the standard pitot tube is used as
part of an assembly, the tube shall be in an
Interference-free  arrangement (subject  to
the approval of the  Administrator).
  4.3 Temperature   Gauges.  After  each
 field  use,  calibrate  dial  thermometers,
 liquid-filled bulb thermometers, thermocou-
 ple-potentiometer   systems,   and   other
 gauges  at a temperature within 10 percent
 of the average absolute stack temperature.
 For temperatures up to 405° C (761° F), use
 an ASTM mercury-in-glass reference  ther-
 mometer, or equivalent, as a reference; al-
 ternatively, either a reference thermocouple
 and potentiometer  (calibrated by NBS) or
 thermometric fixed points, e.g., ice bath and
 boiling water (corrected for barometric pres-
 sure) may be used. For temperatures above
 405° C (761° F), use an NBS-calibrated refer-
 ence thermocouple-potentiometer system or
 an alternate reference,  subject to the ap-
 proval of the Administrator.
  If, during calibration, the absolute tem-
 peratures measured with  the  gauge being
 calibrated  and the  reference  gauge agree
 within  1.5 percent,  the temperature data
 taken in the field shall be considered valid.
 Otherwise, the pollutant emission test shall
 either be considered invalid or adjustments
 (if appropriate) of the test results shall be
 made, subject to the approval of the Admin-
 istrator.
  4.4 Barometer. Calibrate  the barometer
 used against a mercury barometer.

 5.  Calculations

  Carry out calculations, retaining at least
 one extra decimal figure beyond that of the
 acquired data. Round off figures after final
 calculation.                  •,
  5.1 Nomenclature.

A=Cross-sectional area of stack, m2(ft2).
Bua=Water vapor in  the gas stream (from
    Method 5 or Reference Method 4), pro-
    portion by volume.
 CP=Pitot tube coefficient, dimensionless.
Kf=Pitot tube constant,
       Q7 JH |" (g/g-mole) (mm Hg) "I'/'
          secL   (°K)(mmH20)   J
for the metric system and
         _ft_ P( Ib/lb-mole) (in. Hg)-]'«
         sect   (°R)(in. H2O)  "J
for the English system.

Ma—Molecular weight of stack gas, dry basis
   (see Section 3.6) g/g-mole (Ib/lb-mole).
AT,=Molecular  weight  of  stack  gas,  wet
   basis, g/g-mole (Ib/lb-mole).
        Bv,) +18.0 Bu,
                                         B-26

-------
App. A
   Title 40—Protection of Environment
                            Equation 2-5

Ptar=Barometric pressure  at  measurement
   site, mm Hg (in. Hg).
PC = Stack static pressure, mm Hg (in. Hg).

P. -Absolute stack gas pressure, mm Hg (in.
   Hg).
                            Equation 2-6
P,ui=Standard absolute  pressure, 760 mm
   Hg (29.92 in. Hg).
Q«i=Dry volumetric stack gas flow rate cor-
   rected to standard conditions, dscm/hr
   (dscf/hr).
£,= Stack temperature, °C (°P).
71=Absolute stack temperature, °K, (°R).
=273+t for metric
                            Equation 2-7
=460+2. for English
                            Equation 2-8
TM= Standard absolute temperature, 293 °K
   (528° R)
v,= Average  stack gas velocity,  m/sec (ft/
   sec).
AP= Velocity head of stack gas, mm H,O (in.
   HaO).
3,600= Conversion factor, sec/hr.
18.0= Molecular weight  of water, g/g-mole
   (Ib/lb-mole).
5.2 Average stack gas velocity.
                           Equation 2-9
5.3  Average stack gas dry volumetric flow
rate.
 C.a=3,600 U-
                                   »t(i

                         Equation 2-10
6.  Bibliography

  1.  Mark,  L.  S. Mechanical Engineers'
Handbook. New York, McGraw-Hill Book
Co., Inc. 1951.
  2. Perry, J. H. Chemical Engineers' Hand-
book. New York. McGraw-Hill Book Co.,
Inc. 1960.
  3. Shigehara, R. T., W. P. Todd, and W. S.
Smith. Significance of Errors in Stack Sam-
pling Measurements. U.S.  Environmental
Protection Agency, Research Triangle Park,
N.C. (Presented at the Annual Meeting of
the Air Pollution  Control  Association, St.
Louis, Mo., June 14-19,1970.)
  4. Standard Method for Sampling Stacks
for Particulate Matter. In: 1971 Book of
ASTM Standards, Part 23. Philadelphia, Pa.
1971. ASTM Designation D-2928-71.
  5. Vennard, J. K. Elementary Fluid Me-
chanics. New York. John Wiley and Sons,
Inc. 1947.
  6. Fluid  Meters—Their Theory and Appli-
cation. American Society of Mechanical En-
gineers, New York, N.Y. 1959.
  7. ASHRAE Handbook of Fundamentals.
1972. p. 208.
  8. Annual Book of ASTM Standards, Part
26. 1974. p. 648.
  9. Vollaro, R. F. Guidelines for Type S
Pitot Tube Calibration. U.S. Environmental
Protection Agency. Research Triangle Park,
N.C. (Presented at  1st  Annual Meeting,
Source Evaluation Society, Dayton. Ohio,
September 18,1975.)
  10. Vollaro, R. F. A  Type S Pitot Tube
Calibration Study. U.S. Environmental Pro-
tection  Agency,  Emission Measurement
Branch, Research Triangle  Park, N.C. July
1974.
  11. Vollaro, R. F. The  Effects of  Impact
Opening Misalignment on the Value of the
Type S Pitot Tube Coefficient.  U.S. Envi-
ronmental Protection   Agency,  Emission
Measurement Branch,  Research Triangle
Park, N.C. October 1976.
  12. Vollaro, R. F. Establishment of a Bas-
line Coefficient Value for Properly  Con-
structed Type S Pitot Tubes. U.S. Environ-
mental Protection  Agency, Emission Mea-
surement  Branch,  Research Triangle Park
N.C. November 1976.
  13. Vollaro, R. F.  An Evaluation of Single-
Velocity Calibration Technique as a Means
of Determining Type S Pitot Tubes Coeffi-
cient.   U.S.   Environmental    Protection
Agency, Emission Measurement, Branch, Re-
search Triangle Park N.C. August 1975.
  14. Vollaro, R. F. The Use of Type S Pitot
Tubes for the Measurement of Low Veloci-
ties. U.S. Environmental Protection Agency,
Emission  Measurement Branch,  Research
Triangle Park N.C. November 1976.
  15. Smith, Marvin L. Velocity Calibration
of  EPA   Type  Source  Sampling  Probe.
United Technologies Corporation, Pratt and
Whitney Aircraft Division, East Hartford,
Conn. 1975.
  16. Vollaro, R. F.  Recommended Proce-
dure for Sample Traverses in Ducts Smaller
than 12 Inches in  Diameter. U.S. Environ-
mental Protection  Agency, Emission Mea-
surement  Branch,  Research Triangle Park
N.C. November 1976.
  17. Ower,  E.  and R. C. Pankhurst. The
Measurement of Air Flow, 4th  Ed., London,
Pergamon Press. 1966.
  18. Vollaro, R. F. A Survey of Commercial-
ly Available Instrumentation for the Mea-
                                        B-27

-------
Chapter I—Environmental Protection Agency
                                 APP.A
surement of Low-Range Gas Velocities. U.S.
Environmental Protection Agency, Emission
Measurement  Branch,  Research Triangle
Park  N.C. November 1976. (Unpublished
Paper)
  19. Gnyp. A. W., C. C. St. Pierre, D. S.
Smith, D. Mozzon.'and J. Steiner. An Ex-
perimental Investigation of the Effect of
Pitot Tube-Sampling  Probe  Configurations
on the Magnitude of the S Type Pitot Tube
Coefficient  for  Commercially Available
Source Sampling Probes. Prepared  by the
University of Windsor for the  Ministry of
the Environment, Toronto, Canada. Febru-
ary 1975.

METHOD 3—GAS ANALYSIS FOR CARBON DIOX-
  IDE, OXYGEN, EXCESS AIR, AND DRY MOLEC-
  ULAR WEIGHT

1. Principle and Applicability
  1.1  Principle.  A gas sample is extracted
from a stack, by one of the following meth-
ods: (1) single-point, grab  sampling; (2)
single-point, integrated  sampling;  or (3)
multi-point,  Integrated  sampling. The gas
sample is analyzed for percent carbon diox-
ide (COi), percent oxygen (Oa), and, if neces-
sary, percent carbon  monoxide (CO). If a
dry molecular weight determination is to be
made, either an Orsat or a Pyrite * analyzer
may be used for the analysis; for excess air
or emission rate correction factor determi-
nation, an Orsat analyzer must be used.
  1.2  Applicability. This method is applica-
ble for determining CO, and O2 concentra-
tions, excess air, and  dry molecular weight
of a sample from a gas stream of a fossil-
fuel combustion process. The method may
also be applicable to other processes where
it has been  determined that  compounds
  'Mention of trade names or specific prod-
ucts does not constitute endorsement by the
Environmental Protection Agency.
 other than CO*, O2, CO, and nitrogen (n,)
 are not present in concentrations sufficient
 to affect the results.
   Other methods, as well as modifications to
 the procedure described herein, are also ap-
 plicable for some or all of the above deter-
 minations.  Examples  of specific  methods
 and modifications include: (Da multi-point
 sampling method using an Orsat analyzer to
 analyze individual grab samples obtained at
 each point; (2)  a method using CO, or O*
 and stoichiometric calculations to determine
 dry molecular weight and excess air; (3) as-
 signing a value  of 30.0 for  dry  molecular
 weight, in lieu of actual measurements, for
 processes burning natural gas, coal, or oil.
 These  methods  and modifications may be
 used, but are subject to the approval of the
 Administrator, U.S. Environmental Protec-
 tion Agency.

 2. Apparatus
   As an alternative to  the sampling appara-
 tus and  systems described  herein,  other
 sampling systems (e.g., liquid displacement)
 may be used  provided such systems are ca-
 pable  of  obtaining a representative sample
 and maintaining a constant  sampling rate,
 and. are  otherwise capable of yielding ac-
 ceptable results.  Use of such  systems is sub-
 ject to the approval of the Administrator.
   2.1  Grab Sampling (Figure 3-1).
   2.1.1 Probe. The probe should be made of
 stainless  steel or borosilicate  glass tubing
 and should be equipped with an in-stack or
 out-stack filter to remove particulate matter
 (a plug of glass wool is satisfactory for this
 purpose). Any other materials inert  to Oa,
 CO,, CO, and Ni and  resistant to tempera-
 ture at sampling conditions may.be used for
 the probe;  examples of such  material are
..aluminum,.copper, quartz glass and Teflon.
   2.1.2 Pump. A one-way squeeze bulb, or
 equivalent, is used to transport  the gas
 sample to the analyzer.
   2.2  Integrated Sampling (Figure 3-2).
   2.2.1 Probe. A probe  such  as  that de-
 scribed in Section 2.1.1 is suitable.

                                        B-28

-------
 App. A
                        PROBE
               -FILTER (GLASS WOOL)
   Tills 40—Protection of Environment


     FLEXIBLE TUBING

                       /^
                            TO ANALYZER
                                    SQUEEZE BULB
                                 Figure 3-1. Grab-sampling train.
                                                RATE METER
          AIR-COOLED
          CONDENSER
PROBE

    \
        FILTER
     (GLASS WOOL)
                                                              PUMP
                                               QUICK DISCONNECT-


                                           VALVE
                                      RIGID CONTAINER'
                                                                         BAG
                          Figure 3-2. Integrated gas-sampling train.
  2.2.2  Condenser. An air-cooled or water-
 cooled condenser, or other condenser that
 will not remove O,, CO,, CO, and N,, may be
 used  to  remove  excess  moisture which
 would interfere  with the operation of the
 pump and flow meter.
  2.2.3  Valve.  A  needle  valve Is used to
adjust sample gas flow rate.
  2.2.4  Pump.  A leak-free, diaphragm-type
pump,  or  equivalent, is used to transport
sample gas to the  flexible  bag. Install a
small surge tank between the  pump and
                                      B-29

-------
r
                               Chapter I—Environmental Protection Agency
                                 App, A
                               rate meter to eliminate the pulsation effect
                               of the diaphragm pump on the rotameter.
                                 2.2.5  Rate  Meter.  The  rotameter,  or
                               equivalent rate meter, used should be capa-
                               ble of measuring flow rate to within ±2 per-
                               cent of the selected flow rate. A flow rate
                               range of 500 to 1000 cmVmin is suggested.
                                 2.2.6  Flexible Bag. Any leak-free plastic
                               (e.g.. Tedlar. Mylar, Teflon)  or plastic-
                               coated  aluminum (e.g.,  aluminized Mylar)
                               bag) or equivalent, having a capacity con-
                               sistent with the selected flow rate and time
                               length 'Of the test run, may be  used. A ca-
                               pacity in the range of 55 to 90 liters is sug-
                               gested.
                                To  leak-check the  bag, connect it to a
                               water manometer and pressurize the bag to
                               5  to 10 cm H,O (2 to 4 in. H,O). Allow to
                               stand for 10 minutes. Any displacement in
                               the water manometer indicates  a leak. An
                               alternative leak-check method is to pressur-
                               ize the bag to 5 to 10 cm H,O (2 to 4 in. H,O)
                               and allow to stand overnight. A deflated bag
                               indicates a leak.
                                2.2.7  Pressure  Gauge. A water-filled  U-
                               tube manometer, or equivalent, of about 28
                               cm (12 in.) is used for the flexible bag leak-
                               check.
                                2.2.8  Vacuum Gauge. A mercury mano-
                               meter, or equivalent, of at least 760 mm Hg
                               (30 in. Hg) is used for the sampling train
                               leak-check.
                                2.3  Analysis. For Orsat and Fyrite ana-
                               lyzer  maintenance and  operation  proce-
                               dures, follow the instructions recommended
                               by the manufacturer,  unless otherwise spec-
                               ified herein.
                                2.3.1  Dry Molecular Weight Determina-
                               tion. An Orsat analyzer or Fyrite type com-
                               bustion gas analyzer may be used.
                                2.3.2  Emission Rate Correction Factor or
                               Excess Air Determination. An Orsat analyz-
                               er must be used. For low CO, (less than 4.0
                               percent) or high O. (greater that 15.0 per-
                               cent) concentrations, the measuring burette
                               of the Orsat must have at least 0.1 percent
                               subdivisions.

                               3. Drv Molecular Weight Determination
                               Any of the three sampling and analytical
                               procedures described below may be used for
                               determining the dry molecular weight.
                               3.1  Single-Point, Grab   Sampling  and
                               Analytical Procedure.
                               3.1.1  The sampling point  in  the duct
                               shall either be at the centroid of the cross
                               section or at a point no closer to the walls
                               than 1.00 m (3.3 ft), unless otherwise  speci-
                               fied by the Administrator.
                               3.1.2  Set up the equipment as shown  in
                              Figure  3-1, making  sure  all connections
                              ahead of the  analyzer are tight and leak-
                              free. If and Orsat analyzer is used, it is rec-
                              ommended  that  the  analyzer be leaked-
                              checked by following  the procedure in Sec-
                              tion 5; however, the leak-check is optional.
  3.1.3  Place the probe in the  stack, with
 the tip of the probe positioned at the sam-
 pling point; purge the sampling line. Draw a
 sample into the  analyzer and immediately
 analyze it for percent CO, and  percent O>.
 Determine the percentage of the gas that is
 Ni  and CO by subtracting the sum of the
 percent CO3 and percent O» from  100 per-
 cent. Calculate the dry molecular weight as
 indicated in Section 6.3.
  3.1.4  Repeat the sampling, analysis, and
 calculation procedures, until the dry molec-
 ular  weights of  any  three grab  samples
 differ from  their  mean by no more than 0.3
 g/g-mole  (0.3  Ib/lb-mole).  Average,  these
 three molecular weights, and report the re-
 sults to the nearest  0.1 g/g-mole (Ib/lb-
 mole).
  3.2  Single-Point,  Integrated   Sampling
 and Analytical Procedure.
  3.2.1  The sampling point in the  duct
 shall be located as specified in Section 3.1.1.
  3.2.2  Leak-check (optional) the  flexible
 bag as in Section 2.2.6.. Set up  the equip-
 ment as shown in Figure 3-2. Just  prior to
 sampling, leak-check (optional) the  train by
 placing a vacuum gauge  at the condenser
 inlet, pulling a vacuum of at least  250 mm
 Hg (10  in.  Hg), plugging  the outlet at the
 quick disconnect, and then turning off the
 pump. The vacuum should remain stable for
 at least 0.5 minute. Evacuate the  flexible
 bag. Connect the probe and place it in the
 stack, with  the tip of  the probe positioned
 at the sampling point; purge the sampling
 line. Next,  connect the bag and make sure
 that all connections are tight and leak free.
  3.2.3  Sample at a constant rate. The sam-
 pling run should  be simultaneous with, and
 for the same total length of time  as, the pol-
 lutant emission rate determination. Collec-
 tion of at least 30 liters (1.00 ft3) of sample
 gas Is recommended; however, smaller vol-
 umes may be collected, if desired.
  3.2.4  Obtain one  integrated  flue  gas
 sample during each pollutant emission rate
 determination.  Within 8 hours after the
sample is taken, analyze  it for percent CO,
 and percent O, using either an Orsat analyz-
 er or a Fyrite-type combustion gas analyzer.
 If an Orsat analyzer is  used, it is recom-
 mended that the Orsat leak-check described
 in Section 5 be performed before this deter-
mination;  however, the check is optional.
Determine the percentage of the gas that is
Ni and CO  by subtracting' the sum of the
percent CO3 and  percent  O, from 100 per-
cent. Calculate the dry molecular weight as ,
indicated In  Section 6.3.
  3.2.5 Repeat the analysis and  calculation
procedures  until the Individual dry  molecu-
lar  weights for any three analyses differ
from  their  mean  by no more than  0.3 g/g-
mole  (0.3 Ib/lb-mole).  Average these  three
molecular weights, and report the results to
the nearest 0.1 g/g-mole (0.1 Ib/lb-mole).
                                                                      B.-30

-------
APP- A

  3.3  Multi-Point, Integrated Sampling and
Analytical Procedure.
  3.3.1  Unless  otherwise  specified by the
Administrator, a minimum of eight traverse
points shall be  used for  circular  stacks
having diameters less than 0.61 m (24 in.), a
minimum of nine shall be used for rectangu-
lar stacks having equivalent diameters less
than  0.61 m (24 in.), and  a  minimum  of
twelve traverse points shall be used for all
other cases. The traverse points shall be lo-
cated according to Method 1. The  use  of
fewer points is subject to approval  of the
Administrator.
  3.3.2  Follow  the procedures outlined  in
sections 3.2.2 through! 3.2.5, except for the
following: traverse all sampling points and
sample at each point for an equal length of
time.  Record sampling data  as  shown  in
Figure 3-3.

4. Emission Rate  Correction  Factor  or
Excess Air Determination
  NOTE: A Fyrite-type combustion gas  ana-
lyzer is not acceptable for excess air or emis-
   Title 40—Protection of Environment

sion rate correction factor determination,
unless approved  by the Administrator,  if
both percent CO, and percent O, are meas-
ured,  the analytical results of any  of the
three procedures given below may also be
used  for calculating  the  dry.  molecular
weight.
  Each of the three procedures below shall
be used only when specified in an applicable
subpart of the standards. The use of these
procedures for other purposes  must have
specific prior approval of the Administrator.
  4.1  Single-Point,  Grab   Sampling  and
Analytical Procedure.
  4.1.1 The  sampling  point  in the duct
shall either be at the centroid of the cross-
section or at a point no closer to the walls
than 1.00 m (3.3 ft), unless otherwise speci-
fied by the Administrator.
  4.1.2 Set up the equipment as shown in
Figure 3-1,  making  sure  all connections
ahead of the analyzer  are  tight and leak-
free. Leak-check the Orsat analyzer accord-
ing to the procedure described in Section 5.
This leak'Check is mandatory.
TIME




TRAVERSE
PT.




AVERAGE
Q
1pm





%DEV.a





                            avf
                                              (MUST BE < 18%)
                               Figure 3-3. Sampling rate data.
  4.1.3  Place the probe in the stack, with
the tip of the probe positioned at the sam-
pling point; purge the sampling line. Draw a
sample into the analyzer. For emission rate
correction factor determination, immediate-
ly analyze the sample, as outlined in Sec-
tions 4.1.4 and 4.1.5, for percent CO, or per-
cent Oa. If excess air is desired, proceed as
follows: (1) immediately analyze the sample,
as in Sections 4.1.4 and  4.1.5, for percent
COa, Oa, and CO; (2) determine the percent-
age of the gas that is N: by subtracting the
                                        B-31

-------
  Chapter I—Environmental Protection Agency
                                  App. A
  turn of the percent CO>. percent O,, and per-
  cent CO from 100 percent; and (3) calculate
  percent excess air as outlined in Section 6.2.
   4.1.4  To Insure complete  absorption of
  the CO., Oi,  or if applicable, CO, make re-
  peated passes through each absorbing solu-
  tion until two consecutive readings are  the
  same. Several passes (three or four) should
  be  made  between  readings.  (If  constant
  readings cannot be obtained after three con-
  lecutive readings, replace the absorbing so-
  lution.)
   4.1.5 After the analysis  is  completed,
  leak-check (mandatory) the Orsat analyzer
  once again, as described in Section 5. For
  the results of the analysis to be valid, the
  Orsat analyzer must  pass this  leak  test
 before and after the analysis.
   NOTE:  Since this single-point,  grab sam-
 pling and  analytical  procedure in normally
 conducted  in conjunction with  a single-
 point, grab sampling and analytical proce-
 dure for a pollutant,  only one analysis is or-
 dinarily conducted.   Therefore, great care
 must be taken to obtain a valid sample and
 analysis. Although in most cases only CO,
 or O« Is required, It is recommended that
 both COi and O. be measured, and that  Ci-
 tation 5 in the Bibliography be used to vali-
 date the analytical data.
   4.2  Single-Point,   Integrated   Sampling
 and Analytical Procedure.
   4.2.1  The sampling  point  in  the  duct
 •hall be located as specified in Section 4.1.1.
   4.2.2  Leak-check (mandatory)  the flexi-
 ble bag as In Section 2.2.6. Set up the equip-
 ment as shown In Figure 3-2. Just prior to
 sampling, leak-check  (mandatory) the train
 by placing a vacuum gauge at the condenser
 inlet, pulling a vacuum of a least 250 mm Hg
 (10 in. Hg), plugging the outlet at the quick
 disconnect, and then  turning off the pump.
 The vacuum shall remain stable for at least
 0.6 minute. Evacuate  the flexible bag. Con-
 nect the probe and  place it  in the stack,
 with the tip of the probe positioned at the
 sampling point; purge the sampling  line.
 Next, connect the bag and make sure that
 all connections are tight and leak free.
  4.2.3 Sample at a constant rate, or  as
 specified by the  Administrator.  The sam-
 pling run must be simultaneous  with, and
 for the same total lengh of time as, the pol-
 lutant emission rate determination. Collect
 at least  30 liters  (1.00 fts) of sample gas.
 Smaller volumes may be collected, subject
 to approval of the Administrator.
  4.2.4 Obtain one  integrated  flue  gas
sample during each pollutant  emission rate
determination. For emission rate correction
factor determination, analyze the sample
within 4 hours after it is taken for percent
CO. or percent O, (as outlined In Sections
4.2.5  through 4.2.7).  The Orsat  analyzer
must be  leak-check (see Section  5) before
the analysis. If excess air Is desired, proceed
 as follows:  (1) within  4 hours  after the
 sample is taken, analyze it (as in Sections
 4.2.5 through 4.2.7) for percent CO., O,, and
 CO; (2) determine the percentage of the gas
 that is Na by subtracting the sum of the per-
 cent CO,, percent O,, and percent CO from
 100 percent; (3) calculate percent excess air,
 as outlined in Section 6.2.
   4.2.5 To insure  complete  absorption of
 the CO,, O,, or if applicable, CO, make re-
 peated passes through each absorbing solu-
 tion  until two consecutive readings are the
 same. Several passes (three of four) should
 be make between  readings. (If constant
 readings cannot be  obtained after three con-
 secutive readings, replace the absorbing so-
 lution.)
   4.2.6 Repeat the analysis  until the fol-
 lowing criteria are met:
   4.2.6.1   For percent CO.,  repeat the ana-
 lytical procedure until the  results of any
 three analyses differ by no more that (a) 0.3
 percent by volume when CO, is greater than
 4.0 percent or  (b) 0.2 percent by volume
 when CO3 is less than or equal to 4.0 per-
 cent. Average the three acceptable values of
 percent CO, and report the results to the
 nearest 0.1 percent.
   4.2.6.2   For percent Oa, repeat the analyt-
 ical procedure until the results of any three
 analyses differ by no more than (a) 0.3 per-
 cent  by  volume when O, is less than 15.0
 percent or (b) 0.2 percent by volume when
 O. is greater than or equal  to 15.0 percent.
 Average  the three acceptable values of per-
 cent O, and report the results to the nearest
 0.1 percent.
  4.2.6.3   For percent CO, repeat the ana-
 lytical procedure until the results of any
 three  analyses differ by no  more than 0.3
 percent.   Average  the  three  acceptable
 values of percent CO and report the results
 to the nearest 0.1 percent.
  4.2.7 After the  analysis  is  completed,
 leak-check (mandatory) the Orsat analyzer
 once  again, as described in Section 5. For
 the results of the analysis to be valid, the
 Orsat analyzer  must pass  this leak test
 before an after the analysis.

  NOTE: Although in most  Instances  only
 CO, or O, is required, it is recommended
 that both CO, and O, be measured, and that
 Citation  5 in  the Bibliography  be used to
 validate the analytical data.
  4.3  Multi-Point, Integrated Sampling and
 Analytical Procedure.
  4.3.1 Both the minimum number of sam-
 pling  points and the sampling point location
shall be as specified in Section 3.3.1 of this
method. The use of fewer points than speci-
fied is subject to the approval of the Admin-
istrator.
  4.3.2 Follow the  procedures outlined in
Sections 4.2.2 through 4.2.7, except for the
following: Traverse  all sampling points and
                                         R-32

-------
App. A
   Title 40—Protection of Environment
sample at each point for an equal length of
time. Record  sampling  data  as  shown  in
Figure 3-3.

5. Leak-Check Procedure for Orsat Analysers
 Moving  an   Orsat  analyzer  frequently
causes it to leak. Therefore, an Orsat ana-
lyzer should be throughly leak-checked on
site before the flue gas sample is introduced
into it. The procedure for leak-checking an
Orsat analyzer is:
 5.1.1 Bring  the liquid level in each pi-
pette up to the reference mark on the capil-
lary tubing and then close the pipette stop-
cock.
  5.1.2 Raise the leveling bulb sufficiently
to bring the confining liquid meniscus onto
the  graduated  portion  of the burette and
then close the manifold stopcock.
  5.1.3 Record the meniscus position.
  5.1.4 Observe the menicus in the burette
and the liquid level in the pipette for move-
ment over the next 4 minutes.
  5.1.5 For the Orsat analyzer to pass the
leak-check, two conditions must be met.
  5.1.5.1   The liquid  level in  each pipette
must not  fall below the bottom of the capil-
lary tubing during this 4-minute interval.
  5.1.5.2   The meniscus in the burette must
not change by more than 0.2 ml during this
4-minute interval.
  5.1.6 If the analyzer fails the leak-check
procedure, all rubber connections and stop-
cocks should be checked until the cause of
the leak  is identified.  Leaking stopcocks
must be  disassembled, cleaned, and re-
greased.  Leaking rubber connections must
be replaced. After the analyzer is reassem-
bled, the  leak-check procedure must be re-
peated.

6.  Calculations

  6.1 Nomenclature.

j|&=Dry molecular weight, g/g-mole Ub/lb-
    mole).
%EA=Percent excess  air.
%COa=Percent CO2 by volume (dry basis).
%Oj=Pereent O, by volume (dry basis).
%CO=Percent CO by volume (dry basis).
%Ni=Percent Na by volume (dry basis).
0.264=Ratio of Oa to N, in ah-,  v/v.
0.280=Molecular weight of N, or CO, divid-
    ed by  100.
0.320=Molecular weight of O* divided  by
    100.
0.440=Moleeular weight of CO, divided by
    100.

   6.2 Percent Excess Air. Calculate the per-
cent excess air (if applicable), by substitut-
ing the  appropriate  values of  percent O?,
CO, and  N3 (obtained from Section 4.1.3 or
4.2.4) into Equation 3-1.
               %02-0.5%CO	I
                N2 (%02-0.5 %CO) J1UU

                            Equation 3-1
  NOTE: The equation above assumes  that
ambient air is used as the source of O, and
that the fuel does not contain appreciable
amounts of N2 (as do coke oven or blast fur-
nace gases). For those cases when apprecia-
ble amounts of N? are present (coal, oil, and
natural gas do  not  contain  appreciable
amounts of Na) or when oxygen enrichment
is used, alternate  methods, subject to ap-
proval of the Administrator, are required.
  6.3  Dry Molecular Weight. Use Equation
3-2 to calculate the dry molecular weight of
the stack gas
M((=0.440(%COJ)+0.320(%Oa)+
                        0.280(%Na+%CO)
                             Equation 3-2
  NOTE: The above equation does not consid-
er argon in air (about 0.9  percent, molecu-
lars  weight of 37.7). A negative error of
about 0.4  percent is introduced. The tester
may  opt  to  include argon in the analysis
using procedures subject to approval of the
Administrator.                   ,

7. Bibliography         .
  1. Altshuller, A. P.  Storage of  Gases and
Vapors in Plastic Bags. International Jour-
nal of Air and Water Pollution.  6:75-81.
1963.
  2. Conner, William D. and J. S.  Nader. Air
Sampling with Plastice Bags. Journal of the
American  Industrial  Hygiene  Association.
25:291-297. 1964.
  3. Burrell Manual for Gas Analysts, Sev-
enth   edition.  Burrell Corporation,  2223
Fifth Avenue, Pittsburgh, Pa. 15219. 1951.
  4. Mitchell, W.. J. and M.  R. Midgett. Field
Reliability of the Orsat Analyzer. Journal
of Air Pollution Control Association 26:491-
495. May 1976.
  5. Shigehara, R. T., R. M. Neulicht, and
W. S. Smith. Validating Orsat Analysis Data
from Fossil  Fuel-Fired Units. Stack  Sam-
pling News. 4(2):21-26. August, 1976.

  METHOD 4—DETERMINATION OF MOISTURE
          CONTENT IN STACK GASES

1. Principle and Applicability
  1.1   Principle. A gas sample is extracted
at a constant rate from the source; moisture
is removed from the sample stream and de-
termined  either volumetrically or gravime-
trically.
  1.2  Applicability. This method is applica-
ble for determining the moisture content of
stack gas.
  Two procedures are given. The first is a
reference  method, for accurate  determina-
                                        B.-33

-------
Chapter I—Environmental Protection Agency
                                 APP.A
lions  of moisture  content (such  as are
needed  to  calculate  emission  data). The
second is an approximation method, which
provides estimates of percent moisture to
kid in setting isokinetic sampling rates prior
to a pollutant emission measurement run.
The   approximation   method   described
herein is only a suggested  approach;  alter-
native means for approximating the  mois-
ture content, e.g., drying tubes, wet bulb-dry
bulb techniques, condensation  techniques,
stoichiometrlc calculations,  previous experi-
ence, etc., are also acceptable.
  The reference  method is  often conducted
simultaneously  with a  pollutant emission
measurement run; when it is, calculation of
percent isokinetic, pollutant emission rate,
etc., for the run  shall be based upon the re-
sults of the reference  method or its equiva-
lent; these calculations  shall not be based
upon  the  results of the  approximation
method, unless the approximation method
is shown, to the satisfaction of the Adminis-
trator,  U.S.  Environmental  Protection
Agency, to be capable  of  yielding results
within  1  percent H»O  of the  reference
method.

  NOTE: The  reference  method may yield
questionable results when  applied to satu-
rated gas streams or to streams that contain
water droplets. Therefore, when these con-
ditions exist or are suspected, a second de-
termination of the moisture content shall
be made simultaneously with the reference
method, as follows: Assume that the gas
stream  is saturated. Attach a temperature
sensor [capable of measuring to  ±1° C (2*
F)] to the reference method probe. Measure
the stack gas temperature  at each traverse
point (see Section  2.2.1) during the refer-
ence method traverse; calculate the average
stack gas temperature. Next,  determine the
moisture percentage, either by: (1) using a
psychrometric chart and making' appropri-
ate corrections if stack pressure is different
from that of the chart, or  (2) using satura-
tion vapor pressure tables. In cases where
the pyschrometric chart or the saturation
vapor pressure tables  are not applicable
(based on evaluation of the process), alter-
nate methods, subject to the approval of the
Administrator, shall be used.

2. Reference Method
  The procedure  described  in Method 5 for
determining moisture content is acceptable
as a reference method.
  2.1  Apparatus. A schematic of the sam-
pling train used in this reference method is
shown in Figure 4-1. All components shall
be .maintained and calibrated according to
the procedure outlined in Method 5.
                                       B-34

-------
CO
on
                                             FILTER
                                        (EITHER IN STACK
                                        OR OUT OF STACK)
STACK
WALL
CONDENSER-ICE BATH SYSTEM INCLUDING
                SILICA GEL TUBE—j
                                                                                                                     MAIN VALVE
                                                                                                               AIR-TIGHT
                                                                                                                 PUMP
V

>
                                                                                                                                           o
                                                                                                                                           o
                                                                  Figure 4-1. Moisture sampling train-reference method.

-------
 Chapter I—Environmental Protection Agency
                                 APP.A
  2.1.1  Probe. The probe Is constructed of
 stainless steel or glass tubing, sufficiently
 heated to prevent water condensation, and
 Is equipped with  a filter, either in-stack
 (e.g., a plug of glass wool Inserted Into  the
 end of the probe) or heated out-stack (e.g.,
 as described in Method 5), to remove partic-
 ular matter.
  When  stack  conditions  permit,  other
 metals or plastic tubing may be used for  the
 probe, subject to the approval of the Admin-
 istrator.
  2.1.2  Condenser.  The condenser consists
 of four impingers connected  in series with
 ground glass, leak-free fittings or any simi-
 larly  leak-free  non-contaminating fittings.
 The first, third, and fourth impingers shall
 be of the Greenburg-Smith design, modified
 by replacing the tip with a 1.3 centimeter
 (V4 inch) ID glass tube extending to about
 1.3 cm (V4 in.) from the bottom of the flask.
 The second impinger shall be of the Green-
 burg-Smith design  with the standard  tip.
 Modifications  (e.g., using flexible connec-
 tions between the  Impingers, using materi-
 als  other  than glass, or  using flexible
 vacuum lines to connect the filter holder to
 the condenser) may be used, subject to  the
 approval of the Administrator.
  The  first two impingers  shall contain
 known volumes of water, the third shall be
 empty,  and  the fourth  shall  contain a
 known weight of 6- to 16-mesh indicating
 type silica gel, or equivalent desiccant. If
 the silica gel has been previously used,  dry
 at 175° C (350° P) for 2 hours. New silica gel
 may be used as received. A thermometer, ca-
pable of measuring temperature to within 1°
 C (2* F), shall be placed at the outlet of  the
 fourth impinger, for monitoring purposes.
  Alternatively,  any system  may  be  used
 (subject to the approval of the Administra-
tor) that cools  the  sample gas stream and
 allows measurement of both the water that
has been condensed and the moisture leav-
ing the condenser, each to within 1 ml or 1
g. Acceptable means are to measure the con-
densed water, either gravlmetrically or volu-
metrically, and  to  measure  the moisture
leaving the condenser by: (1) monitoring  the
temperature and pressure at the exit of  the
condenser and using Dalton's law of partial
pressures,  or (2) passing the sample  gas
stream through a tared silica gel (or equiva-
lent desiccant)  trap, with exit gases kept
below 20' C (68' P),  and determining the
weight gain.
  If means other than silica gel are used to
determine the amount of moisture leaving
the condenser, it Is recommended that silica
gel (or equivalent) still be used between  the
condenser  system  and pump,  to prevent
moisture condensation In the pump and me-
tering devices and to avoid the need to make
corrections for  moisture  in the  metered
volume.
  2.1.3  Cooling System. An ice bath  con-
tainer and crushed ice (or  equivalent) are
used to aid in condensing moisture.
  2.1.4  Metering System. This  system in-
cludes a  vacuum gauge, leak-free  pump,
thermometers  capable  of measuring  tem-
perature to within 3° C (5.4° P), dry gas
meter capable  of measuring  volume  to
within 2 percent, and related equipment as
shown in  Figure 4-1.  Other metering sys-
tems,  capable  of maintaining  a  constant
sampling rate and determining sample gas
volume, may be used, subject to the approv-
al of the Administrator.
  2.1.5  Barometer. Mercury,  aneroid,  or
other barometer capable of measuring at-
mospheric  pressure to  within 2.5 mm Hg
(0.1 in. Hg) may be used. In many cases, the
barometric reading may be obtained from a
nearby national weather service station, in
which case the station value (which is the
absolute barometric pressure) shall  be re-
quested and an adjustment for elevation dif-
ferences between the weather station and
the sampling point shall be applied at a rate
of minus 2.5 mm Hg (0.1 in. Hg) per 30 m
(100 ft) elevation increase or vice versa for
elevation decrease.
  2.1.6  Graduated  Cylinder and/or  Bal-
ance. These items are used to measure con-
densed water  and moisture caught in the
silica gel to within 1 ml or 0.5 g. Graduated
cylinders shall have subdivisions no greater
than 2 ml. Most laboratory balances are ca-
pable of weighing to the nearest 0.5  g  or
less. These balances  are  suitable for use
here.
  2.2  Procedure. The  following procedure
is written for  a condenser system (such as
the impinger  system described  in Section
2.1.2) incorporating volumetric analysis  to
measure the condensed moisture, and silica
gel and gravimetric analysis  to measure the
moisture leaving the condenser.
  2.2.1 Unless  otherwise specified  by the •
Administrator, a minimum of eight traverse
points shall  be  used for  circular  stacks
having diameters less than 0.61 m (24 in.), a
minimum  of nine points shall be used for
rectangular stacks having equivalent diame-
ters less than 0.61 m (24 in.), and a mini-
mum of twelve traverse points shall be used
in all other cases. The traverse points shall
be located according to Method  1. The use
of fewer points is subject to  the approval of
the Administrator. Select a suitable probe
and  probe length such that all traverse
points can be sampled. Consider sampling
from  opposite sides of  the stack (four total
sampling ports) for large stacks,  to permit
use of  shorter probe  lengths.  Mark the
probe with heat resistant tape  or by some
other method to denote the  proper distance
into the stack or duct for  each sampling
point. Place known volumes  of water in the
first two impingers. Weigh  and record the
                                      B-36

-------
 App. A

 weight of the silica gel to the nearest 0.5 g,
 and transfer the silica gel to the fourth im-
 pinger; alternatively, the silica gel may  first
 be transferred  to the impinger, and  the
 weight of the silica gel plus impinger record-
 ed.
   2.2.2  Select a total sampling time such
 that  a minimum total gas volume  of  0.60
 scm (21 scf) will be collected, at a rate no
 greater than 0.021 mVmin (0.75  cfm). When
 both moisture content and pollutant emis-
 sion rate are to be determined, the moisture
 determination shall be simultaneous with,
 and for the same total length of  time as, the
 pollutant  emission rate run, unless  other-
 wise specified in an  applicable  subpart of
 the standards.
  2.2.3  Set up the sampling train as shown
 in Figure 4-1. Turn on the probe heater and
 (if applicable) the  filter heating system to
 temperatures of  about 120° C (248°  P), to
 prevent water condensation ahead  of  the
 condenser; allow  time for the temperatures
 to stabilize. Place crushed ice in the ice bath
 container.  It is  recommended, but not re-
 quired, that a leak check be done, as follows:
 Disconnect the  probe from the first  im-
 pinger  or  (if applicable)  from  the filter
 holder.  Plug  the  inlet to the first impinger
 (or filter holder)  and  pull a 380 mm (15  in.)
 Hg vacuum; a lower vacuum may be used,
provided that it  is not exceeded during  the
 test. A leakage rate in excess of 4 percent of
 the average sampling rate  or 0.00057  mV
min (0.02 cfm), whichever is less,  is unaccep-
table. Following the leak check, reconnect
the probe to the sampling train.
  2.2.4  During the sampling run, maintain
a sampling rate  within 10 percent of con-
   Till* 40—Protection of Environment

 stant rate, or as specified by the Adminis-
 trator.  For each  run, record the  data re-
 quired on the example data sheet shown to
 Figure  4-2. Be sure to record the dry gas
 meter reading at the beginning and end of
 each sampling time  increment and  when-
 ever sampling is halted. Take other appro-
 priate readings at each sample  point, at
 least once during each time increment.
  2.2.5  To  begin  sampling,  position  the
 probe tip at the first traverse point. Imme-
 diately  start the pump and adjust the flow
 to the desired rate. Traverse the cross sec-
 tion, sampling at each traverse point for an
 equal length of time. Add more ice and, if
 necessary, salt to maintain a temperature of
 less  20° C (68° F) at the silica gel outlet.
  2.2.6  After  collecting the sample, discon-
 nect the probe  from the filter holder (or
 from the first impinger) and conduct a leak
 check (mandatory) as described in Section
 2.2.3. Record the leak rate. If the leakage
 rate exceeds the allowable rate, the  tester
 shall either reject the test results or shall
 correct the sample volume as in Section 6.3
 of Method 5. Next, measure the volume of
 the  moisture condensed to the nearest ml.
 Determine the increase in weight of  the
 silica gel (or silica  gel plus impinger) to the
 nearest 0.5 g.  Record this information (see
 example data  sheet, Figure 4-3) and calcu-
 late  the moisture percentage, as described in
 2.3 below.
  2.3 Calculations. Carry out the following
calculations, retaining at least one  extra
decimal figure beyond that of the acquired
data. Round off figures after final calcula-
tion.
                                        B-37

-------
I
                                                                          ;ecAnw	
                                                                          OPERATOR	
                                                                          BATf	
                                                                          RUN NO.	
                                                                          AMIIEKT TEMPERATURE.
                                                                          IAROMETRIC PRESSURE-
                                                                          moiE LENGTH n(W	
                                                                                                                     SCHEMATIC OF STACK CROSS SECTION
                    CO
                    00
TRAVERSE POINT
NUMIER















TOTAL
SAMPLING
TIME
(ei.min.
















AVERAGE
STACK •
TEMPERATURE
•C(«F|

















PRESSURE
DIFFERENTIAL
ACROSS
ORIFICE METER
(AH),
•"•(in.) H{0

















METER
READING
GAS SAMPLE
VOLUME
m) Ift3)

















AVm
»'(tt3)

















GAS SAMPLE TEMPERATURE
AT DRY GAS METER
INLET
(Tmj.l.'Ct'F)















An.
A*
OUTLET
(TiiMtl.'CCF)















**

TEMPERATURE
OF GAS
LEAVING
CONDENSER OR
LAST IMPINGER.
•C(«F)

















                                                                                                     Figure 4-2. Field moisture determination-reference method.
I
i
I
\
a
5*
Jr

-------
APR. A

FINAL
INITIAL
DIFFERENCE
IMPINGE;:
VOLUME.
ml '



SILICA GEL
WEIGHT.
9



     Figure 4-3. Analytical data • reference method.
   Title 40—Protection of Environment

where:
_K\=0.001333 m'/ml for metric units
   =0.04707 ftVml for English units
2.3.3  Volume of water vapor  collected in
   silica gel.
2.3.1  Nomenclature.
B „=Proportion of water vapor, by volume,
    in the gas stream.
Af«,=Molecular weight of water,  18.0 g/g-
    mole (18.0 Ib/lb-mole).
/>»,=Absolute pressure (for this method,.
    same as barometric pressure) at the dry
    gas meter, mm Hg (in. Hg).
P,,,j=Standard  absolute pressure, 760 mm
    Hg (29.92 in. Hg).
j£=Ideal gas  constant, 0.06236 (mm Hg)
    (m3)/(g-mole) (°K) for metric units and
    21.85 (in. Hg) (fts)/(lb-mole> (°R) for
    English units.
T«= Absolute temperature at  meter, °K
    CR).
T.w=Standard absolute  temperature, 293°
    K (528"R).
Vw=Dry gas volume measured by dry gas
    meter, dcm (dcf).
A Vm=Incremental dry gas volume measured
    by dry gas meter at each traverse point,
    dcm (dcf).
V»<«=Volume of water vapor condensed
    corrected  to standard  conditions,  scm
    (scf).
V«.f<«d>=Voiume of water vapor collected in.
    silica gel corrected to  standard  condi-
    tions, scm (scf).
V/=Pinal volume of condenser water,  ml.
V,=Initial  volume,  if any,  of condenser
    water, ml.
Wi=Final weight of silica  gel  or silica gel
    • plus impinger, g.
 Wi=Initial weight of silica gel or silica gel
    plus impinger, g.
 Y =Dry gas meter calibration factor.
p«=Density of water, 0.9982 g/ml (0.002201
    Ib/ml).
2.3.2 Volume of water vapor condensed.
              (Vr-Vi)i>»RTmt.t
                           Equation 4-2
where:
K,=0.001335 m»/g for metric units
   =0.04715 ftVg for English units
2.3.4  Sample gas volume.
                  VmP*
                            Equation 4-3
 where:
 JT,=0.3858 °K/mm Hg for metric units
   =17.64 °R/in. Hg for English units
  NOTE:  If the  post-test lead rate (Section
 2.2.6) exceeds the allowable rate, correct the
 value of Vm in Equation 4-3, as described in
 Section 6.3 of Method 5.
  2.3.5  Moisture Content.
_

 •"*
              Vice («td) + Vv it (»td)      .

            c <«td) + Vir., (ltd) + V« (•")
                             Equation 4-4
                            Equation 4-1
   NOTE: In saturated or moisture droplet-
 laden gas streams,  two calculations  of the
 moisture content of the stack gas shall be
 made,  one using a  value  based upon the
 saturated conditions (see Section 1.2), and
 another based upon the results of the im-
 pinger analysis.  The  lower of these  two
 values of B „ shall be considered correct.
   2.3.6 Verification of constant sampling
 rate. For each time increment, determine
 the AVm. Calculate the average. If the value
                                          B-39

-------
 Chapter I—Environmental Protection Agency
                                 APP.A
 tor any  time Increment differs from the
 average by more than 10 percent, reject the
 results and repeat the run.

 3. Approximation Method
  The  approximation  method described
 below  Is  presented  only as  a suggested
 method (see Section 1.2).
  3.1  Apparatus.
  3.1.1 Probe. Stainless steel  glass tubing,
 sufficiently heated to prevent water conden-
 sation and equipped with a filter (either in-
 stock or heated out-stack) to remove partlc-
 ulate matter. A plug of glass wool, inserted
 into the end  of the probe, is a satisfactory
 filter.
  3.1.2 Impingers. Two midget impingers,
 each with 30 ml capacity, or equivalent.
  3.1.3  Ice Bath. Container  and ice, to aid
 in condensing moisture in impingers.
  3.1.4  Drying Tube.  Tube packed  with
new or regenerated 6- to 16-mesh indicating-
type silica gel (or equivalent desiccant), to
dry the sample gas and to protect the meter
and pump.
  3.1.5  Valve. Needle valve, to regulate the
sample gas flow rate.
  3.1.6  Pump. Leak-free, diaphragm  type,
or  equivalent, to  pull  the  gas  sample
through the train.
  3.1.7  Volume Meter. Dry gas meter, suffi-
ciently accurate  to  measure  the  sample
volume within 2%, and calibrated over the
range of flow rates and conditions actually
encountered during sampling.
  3.1.8  Rate Meter. Rotameter, to measure
the flow range from  0 to 3 1pm (0 to 0.11
cfm).
  3.1.9  Graduated Cylinder. 25 ml.
  3.1.10  Barometer.  Mercury, aneroid,  or
other  barometer,  as  described in Section
2.1.5 above.
  3.1.11  Vacuum Gauge. At least 760 mm
Hg (30 in. Hg) gauge, to be used for the sam-
pling leak check.
  3.2  Procedure.
  3.2.1  Place exactly  5 ml distilled water in
each Impinger.

Leak  check the sampling train as follows:
Temporarily insert a  vacuum  gauge  at  or
near the probe inlet; then, plug the  probe
inlet and pull a vacuum of at least 250 mm
Hg (10  in. Hg).  Note; the time rate  of
change of the dry gas meter dial; alterna-
tively, a rotameter (0-40 cc/min) may  be
temporarily attached  to the  dry gas  meter
outlet to determine the leakage rate. A leak
rate not In excess of  2 percent of the aver-
.age sampling rate is acceptable.

  NOTE: Carefully release  the probe inlet
plug before turning off the pump.
                                      B-40

-------
                                   HEATED PROBE

SILICA GEL TUBE
RATE METER,



     VALVE
co
                                 FILTER

                                 (GLASS WOOL)
                                    ICE BATH
                                     MIDGET IMPINGERS
             PUMP
                                          Figure 4-4. Moisture-sampling train • approximation method.
                                                            3
                                                            3T


                                                            f
                                                            8.

                                                            IW

-------
                                  LOCATION.
                                  TEST
COMMENTS
                                  DATE
                                  OPERATOR
                                  BAROMETRIC PRESSURE
                            5
                            a
                            •a
                            m
                            3
                            I
                            I
-p.
ro
CLOCK TIME





GAS VOLUME THROUGH
METER. (Vm).
m3 (ft3)





RATE METER SETTING
m3/min. (ft3/min.)





METER TEMPERATURE.
°C (°F)





Piniirn 4_R Pialrl mrhic+ura rla+Armina+i/\n _ or\nr/\vimat!nn nnathrkrl
1
to
3
•3
•o

-------
App. A
   Title 40—Protection of Environment
Figure 4-5. Field moisture determination-
approximation method. .

  3.2.2  Connect the probe, insert it into the
stack, and sample at a constant rate of 2
1pm (0.071  cfm). Continue sampling until
the dry gas meter registers about 30 liters
(1.1 ft9) or until visible liquid .droplets are
carried over from the first impinger to the
second. Record temperature, pressure, and
dry  gas  meter readings  as  required  by
Figure 4-5.
  3.2.3  After  collecting the sample, com-
bine the contents of the two impingers and
measure the volume to the nearest 0.5 ml.

  3.3  Calculations. The calculation method
presented is designed to estimate the mois-
ture in the stack gas; therefore, other data,
which are only necessary for accurate mois-
ture determinations, are not collected. The
following equations adequately estimate the
moisture content, for the purpose  of deter-
mining isokinetic sampling rate settings.

  3.3.1  Nomenclature.

B«™=Approximate proportion,  by volume,
   of water vapor in the gas stream leaving
   the second impinger, 0.025.
Bwt=Water vapor in the gas stream, propor-
   tion by volume.
Aft=Molecular weight of water,  18.0  g/g-
   mole (18.0 Ib/lb-mole).
Pm= Absolute  pressure  (for this  method,
   same as barometric pressure) at the dry
   gas meter.
Pii=Final volume of impinger contents, ml.
V/=Initial volume of impinger contents, ml.
V«= Dry gas volume measured  by dry  gas
   meter, dcm (def).
Vm(«d)=Dry gas volume measured by dry gas
   meter, corrected to standard conditions,
   dscm (dscf).
V»c<«=Volume of water vapor condensed,
   corrected to standard  conditions,  scm
   (scf).
p«=Density of water, 0.9982 g/ml (0.002201
   Ib/ml).
Y=Dry gas meter calibration factor.

  3.3.2  Volume of water vapor collected.
where:                             \
        _(Vf-V,)p,rKT.tl
     ' «•'.—     s   »7
        = Kl(Vf-Vi)
                       Kquation 4-5
K,=0.001333 m'/ml for metric units
   =0.04707 ftVml for English units.
  3.3.3  Gas volume.
               V.P.
                        Kquation 4-(i
where:
#,=0.3858 °K/mm Hg for metric units
   = 17.64 °R/in. Hg for English units
  3.3.4  Approximate moisture content.

                  r "*m
                           Equation 4-7
4. Calibration
  4.1  For the reference method, calibrate
equipment as specified in the following sec-
tions  of Method 5: Section 5.3 (metering
system); Section 5.5 (temperature gauges);
and Section  5.7 (barometer). The recom-
mended leak  check of the metering system
(Section 5.6 of Method 5) also applies to the
reference method.  For  the approximation
method, use the procedures outlined in Sec-
tion 5.1.1 of Method 6 to calibrate the me-
tering system, and the procedure of Method
5, Section 5.7  to calibrate the barometer.
5. Bibliography
  1.  Air  Pollution Engineering  Manual
(Second Edition). Danielson, J. A. (ed.). U.S.
Environmental Protection Agency, Office of
Air Quality Planning  and  Standards. Re-
search Triangle Park, N.C. Publication No.
AP-40. 1973.
  2. Devorkin, Howard,  et al. Air Pollution
Source Testing Manual. Air Pollution Con-
trol District,  Los Angeles, Calif. November,
1963.
  3. Methods  for Determination of Velocity,
Volume Dust and  Mist Content of Gases.
Western Precipitation Division of Joy Man-
                                        B-43

-------
Chapter I—Environmental Protection Agency
                                 App. A
ufocturing Co., Los Angeles, Calif. Bulletin
WP-50.1968.
 METHOD ^DETERMINATION OF PARTICULATE
  ISitissibNS FROM STATIONARY SOURCES
1. Principle and Applicability
 1.1 Principle. Particulate matter is with-
drawn isokinetically from the source and
collected on  a glass fiber filter maintained
tt a temperature in the range of 120±14° C
(248±25* F)  or such other temperature as
specified by  an applicable subpart  of  the
standards  or approved  by Administrator,
OJS. Environmental  Protection Agency, for
* particular application.  The  participate
mass, which includes any material that con-
denses at or above the  filtration tempera-
ture, is determined gravimetrically after re-
moval of uncombined water.
 1.2 Applicability. This method is applica-
ble  for  the  determination of  particulate
emissions from stationary sources.
2. Apparatus
  2.1  Sampling Train. A schematic of the
sampling train used in this method is shown
in Figure 5-1. Complete construction details
are given in APTD-0581 (Citation 2 in Sec-
tion 7); commercial models of this train are
also available.  For changes from APTD-
0581 and for allowable modifications of the
train shown in Figure 5-1, see the following
subsections.
  The  operating  and  maintenance proce-
dures for the sampling train are described in
APTD-0576 (Citation 3 in Section 7). Since
correct usage is important in obtaining valid
results, all users should read  APTD-0576
and adopt  the  operating and maintenance
procedures outlined in it, unless otherwise
specified herein.  The  sampling train con-
sists of the following components:
                                       B-44

-------
                                    TEMPERATURE SENSOR
                                            PROBE
                                           TEMPERATURE
                                              SENSOR
                                             /•r
                                                                       IMPINGER TRAIN OPTIONAL, MAY BE REPLACED
                                                                            BY AN EQUIVALENT CONDENSER
                                                                                *
                                                                                •o
                          HEATED AREA
 PITOTTUBE

     PROBE
THERMOMETER

FILTER HOLDER
en
i
en
REVERSE-TYPE
 PITOT TUBE
                                                         THERMOMETER
                                           PITOT MANOMETER

                                                   ORIFICE
                                  IMPINGERS               ICE BATH
                                           BY-PASS VALVE
                                         THERMOMETERS
                                                                                          VACUUM
                                                                                          GAUGE
                                                                                   MAiN VALVE
                                                     DRY GAS METER      AIR-TIG.HT
                                                                         PUMP
                                                     Figure 5-1. Particulate-samplinK train
CHECK
VALVE
                                                                                                       VACUUM
                                                                                                         LINE
                                                                                 HI

                                                                                 I
                                                                                I

-------
Chapter I—Environmental Protection Agency
                                 App. A
  2.1.1  Probe  Nozzle. Stainless steel (316)
or glass With sharp, tapered leading edge.
The  angle  of  taper shall be 30°  and the
taper shall be  on the outside to preserve a
constant internal  diameter. The  probe
nozzle shall be of the button-hook  or elbow
design, unless otherwise specified by the Ad-
ministrator. If made of stainless steel, the
nozzle  shall be constructed from  seamless
tubing; other materials of construction may
be used, subject to the approval of the Ad-
ministrator.
  A range of nozzle sizes suitable for  isokin-
etic sampling should be available,  e.g., 0.32
to 1.27 cm <% to %  in.)—or larger if  higher
volume sampling trains are used—inside di-
ameter (ED) nozzles in increments of 0.16 cm
(%• in.). Each  nozzle shall be calibrated ac-
cording to  the  procedures outlined in Sec-
tion 5.
  2.1.2  Probe  Liner. Borosilicate or  quartz
glass tubing with a heating system capable
of maintaining a gas temperature at the exit
end during sampling of 120±14° C (248±25°
P), or such other temperature as  specified
by an applicable subpart of the standards or
approved by the Administrator for a partic-
ular application. (The tester may opt to op-
erate the equipment at a temperature lower
than that specified.) Since the actual tem-
perature at the outlet of the probe Is not
usually monitored during sampling,  probes
constructed according to APTD-0581 and
utilizing  the calibration curves of APTD-
0576 (or calibrated  according to the proce-
dure outlined in APTD-0576) will be  consid-
ered acceptable.
  Either borosllicate or quartz glass probe
liners may be used for stack temperatures
up to  about 480' C (900' P) quartz liners
shall be used for  temperatures between 480
and 900* C (900 and 1,650° F). Both types of
liners may be used at higher temperatures
than specified for short periods of time, sub-
ject  to the approval of the Administrator.
The softening temperature for borosilicate
is 820° C (1,508° P), and for quartz it is
1,500° C (2,732° P).
  Whenever practical, every  effort  should
be made to use borosilicate or quartz glass
probe  liners.   Alternatively,  metal  liners
(e.g., 316 stainless steel, Incoloy 825,* or
other  corrosion resistant metals) made of
seamless tubing may be used, subject to the
approval of the Administrator.
  2.1.3  Pitot Tube. Type S, as described in
Section 2.1 of Method 2, or other device ap-
proved by the Administrator. The pltot tube
shall be attached to the probe (as shown in
Figure 5-1) to allow constant monitoring of
the  stack  gas velocity.  The impact (high
pressure) opening plane of  the pitot tube
  •Mention of trade names or specific prod-
 uct does not constitute endorsement by the
 Environmental Protection Agency.
shall be even with or above the nozzle entry
plane (see Method 2,  Figure  2-6b)  during
sampling. The Type S pitot tube assembly
shall have a known coefficient, determined
as outlined In Section 4 of Method 2.
  2.1.4 Differentia  Pressure  Gauge.   In-
clined  manometer  or  equivalent  device
(two), as described in Section 2.2 of Method
2. One manometer shall be used or velocity
head (Ap) readings, and the other, for orifice
differentia pressure readings.
  2.1.5 Filter Holder.  Borosilicate glass,
with a glass frit filter support and a silicone
rubber gasket. Other materials of construc-
tion (e.g., stainless steel, Teflon, Viton) may
be used, subject to .approval of the Adminis-
trator. The holder design shall provide  a
positive seal against leakage from the out-
side or around the filter. The holder shall
be attached immediately at the outlet of the
probe (or cyclone, it used).
  2.1.6  Filter Heating System. Any heating
system capable of maintaining a  tempera-
ture around the  filter holder during  sam-
pling of 120±14° C (248±25° F), or  such
other temperature as specified by  an appli-
cable subpart of  the standards or  approved
by the Administrator for a particular appli-
cation. Alternatively, the tester may opt to
operate  the equipment  at a temperature
lower  than that specified. A temperature
gauge capable of measuring temperature to
within 3° C (5.4° P) shall  be installed so that
the temperature around the  filter holder
can be regulated  and monitored during sam-
pling.  Heating systems other  than the one
shown in APTD-05'81 may be used.
   2.1.7  Condenser.  The  following  system
shall be used to determine the stack gas
moisture content: Four implngers connected
in series with leak-free ground glass fittings
or any similar leak-free  non-contaminating
fittings.  The first,  third, and  fourth im-
pingers  shall be of the Greenburg-Smith
design, modified  by replacing the tip with
1.3 cm (Va In.) ID glass tube extending to
about 1.3 cm (V4 in.) from the  bottom of the
flask. The second, impinger shall be of the
Greenburg-Smith design with the standard
tip.  Modifications (e.g.,  using flexible  con-
nections between the impingers, using mate-
rials  other  than glass,  or using  flexible
vacuum lines to connect the filter holder to
the condenser) may be used, subject to the
approval of the Administrator. The first and
second Implngers shall contain known quan-
tities of water (Section 4.1.3), the third shall
be empty, and the fourth shall  contain a
known weight of silica gel,  or  equivalent
desiccant. A thermometer, capable of meas-
uring temperture to within 1° C (2° F) shall
be placed at the outlet of the fourth im-
pinger for monitoring purposes.
   Alternatively,  any system that  cools the
sample gas stream and allows measurement
 of the water condensed and moisture leav-
                                          B-46

-------
App. A
   Title 40—Protection of Environment
ing the condenser, each to within 1 ml or 1 g
may be used, subject to the approval of the
Administrator.  Acceptable  means are  to
measure  the condensed water either  gravi-
metrically or volumetrically and to measure
the moisture leaving the  condenser by: (1)
monitoring the temperature and pressure at
the exit of the condenser and using Dalton's
law of partial pressures; or (2) passing the
sample has stream through a tared silica gel
(or equivalent desiccant) trap with exit
gases kept below 20° C (68° P) and determin-
ing the weight gain.
  If means other  than  silica gel are used to
determine the  amount of moisture leaving
the condenser, it is recommended that silica
gel (or equivalent) still be used between the
condenser  system and pump to prevent
moisture condensation  in the pump and me-
tering devices and to avoid the need to make
corrections  for moisture  in  the  metered
volume.
  NOTE: If  a determination of the partieu-
late matter collected in the impingers is de-
sired in  addition to moisture content, the
impinger system  described above  shall be
used,  without  modification.  Individual
States  or control  agencies requiring this in-
formation  shall  be contacted  as to the
sample recovery  and  analysis of the im-
pinger contents.
  2.1.8  Metering System.  Vacuum  gauge,
leak-free pump,  thermometers capable  of
measuring temperature to within 3° C (5.4°
F), dry  gas meter capable  of measuring
volume to within 2 percent,  and related
equipment, as shown in Figure 5-1. Other
metering systems capable of  maintaining
sampling rates within 10 percent of isokine-
tic and of determining sample volumes  to
within 2 percent may be used, subject to the
approval of the  Administrator. When the
metering system is used in conjunction with
a pitot tube, the system shall enable checks
of isokinetic rates.
  Sampling  trains utilizing  metering sys-
tems designed .for higher flow rates than
that decribed in APTD-0581 or APDT-0576
may be  used provided that  the specifica-
tions of this method are met.
  2.1.9  Barometer.  Mercury  aneroid,  or
other barometer  capable  of  measuring at-
mospheric  pressure to  within 2.5 mm Hg
(0.1 in. Hg). In many cases the barometric
reading may be obtained from a nearby na-
tional weather service station, in which case
the station value (which is  the  absolute
barometric pressure) shall be requested and
an adjustment  for elevation differences be-
tween  the weather  station  and  sampling
point shall be applied at a rate of minus 2.5
mm Hg (0.1 in. Hg) per 30 m (100  ft)  eleva-
tion increase or vice versa for elevation de-
crease.
  2.1.10  Gas Density Determination Equip-
ment.  Temperature sensor  and,  pressure
gauge, as described in Sections 2.3 and 2.4 of
Method 2, and gas analyzer, if necessary, as
described  in Method 3. The temperature
sensor shall, preferably, be permanently at-
tached to the pitot tube or sampling probe
in a fixed configuration, such that the,tip of
the sensor extends beyond the leading edge
of the probe sheath and does not touch any
metal. Alternatively, the sensor may be at-
tached just  prior to use in the  field. Note,
however, that if the temperature sensor is
attached in the field, the sensor must be
placed in  an interference-free arrangement
with respect to the Type S pitot tube open-
ings (see Method 2, Figure 2-7). As a second
alternative, if a difference of not more than
1 percent in the average,velocity measure-
ment is to be introduced, the temperature
gauge need not be attached to the probe or
pitot tube. (This alternative is subject to the
approval of the Administrator.)
  2.2  Sample  Recovery.   The  following
items are needed.
  2.2.1 Probe-Liner    and    Probe-Nozzle
Brushes. Nylon bristle brushes  with stain-
less steel wire handles. The probe  brush
shall have extensions (at least as long as the
probe) of stainless steel. Nylon, Teflon, or
similarly inert  material.  The brushes shall
be properly sized and shaped to brush  out
the probe liner and nozzle.
  2.2.2 Wash  Bottles—Two. Glass  wash
bottles  are  recommended;  polyethylene
wash  bottles may be used at the option of
the tester. It is recommended that acetone
not be stored  in polyethylene  bottles  for
longer than a month.
  2.2.3 Glass Sample  Storage  Containers.
Chemically resistant, borosilicate glass bot-
tles, for acetone washes, 500 ml or 1000 ml.
Screw cap liners  shall  either  be  rubber-
backed Teflon or shall be constructed so as
to be leak-free and resistant to chemical
attack by acetone. (Narrow mouth glass bot-
tles have  been found to be less prone to
leakage.) Alternatively, polyethylene bottles
may be used.
  2.2.4 Petri Dishes.  For  filter samples,
glass or polyethylene, unless otherwise spec-
ified by the Administrator.
  2.2.5 Graduated  Cylinder  and/or Bal-
ance. To measure condensed water to within
1 ml or 1 g. Graduated cylinders shall have
subdivisions no greater than 2 ml. Most lab-
oratory balances are capable of weighing to
the nearest  0.5 g or less. Any of these bal-
ances is suitable or use here and in Section
2.3.4.
  2.2.6 Plastic   Storage  Containers. Air-
tight containers to store silica gel.
  2.2.7 Funnel and Rubber  Policeman. To
aid  in transfer of silica gel to container; not
necessary if silica gel is weighed in the field.
  2.2.8 Funnel. Glass or polyethylene, to
aid in sample recovery.
                                        B-47

-------
Chapter I—Environmental Protection Agency
                                 APR. A
  2.2  Analysis. For analysis, the following
equipment is needed.
  2.3.1 Glass Weighing Dishes.
  2.3.2 Desiccator.
  2.3.3 Analytical Balance.  To measure to
within 0.1 mg.
  2.3.4 Balance. To measure to within 0.5 g.
  2.3.5 Beakers. 250 ml.
  2.3.6 Hygrometer. To measure the rela-
tive  humidity of  the laboratory environ-
ment.
  2.3.7 Temperature  Gauge.  To  measure
the temperature of the laboratory environ-
ment.
3. Reagents
  3.1  Sampling. The reagents used in sam-
pling are as follows:
  3.1.1 Filters. Glass fiber  filters, without
organic binder, exhibiting at least 99.95 per-
cent efficiency <<0.05 percent penetration)
on 0.3-mlcron dioctyl phthalate smoke parti-
cles. The filter efficiency test shall be con-
ducted in accordance  with ASTM standard
method D 2986-71. Test data from the sup-
plier's quality control program are suffi-
cient for this purpose. In sources containing
SO, or SO,, the filter material must be of a
type that is unreactlve to SO, or SO,. Cita-
tion  10 in Section 7 may be used' to select
the appropriate filter.
  3.1.2 Silica Gel. Indicating type,  6 to 16
mesh. If previously used, dry at 175° C (350°
F) for 2 hours. New silica gel may be used as
received. Alternatively, other types of desic-
cants (equivalent or better) may be used,
lubject to the approval of the Administra-
tor.
  3.1.3 Water. When analysis of the mate-
rial caught in the impingers  is required, dis-
tilled water shall be used. Run blanks prior.
to field use to eliminate a high blank on test
samples.
  3.1.4 Crushed Ice.
  3.1.5 Stopcock Grease. Acetone-insoluble,
heat-stable silicone grease. This is not neces-
sary if screw-on connectors  with Teflon
sleeves, or similar, are used. Alternatively,
other types of stopcock grease may be used,
subject to the approval of the Administra-
tor.
  3.2  Sample  Recovery.  Acetone-reagent
grade, <0.001 percent residue, in glass bot-
tles—is required. Acetone  from  metal  con-
tainers generally has  a high residue blank
and should not be used. Sometimes, suppli-
ers transfer acetone to glass bottles from
metal containers; thus, acetone blanks shall
be run prior to field use and only acetone
with low blank values (<0.001 percent) shall
be used. In  no case shall a blank value of
greater than 0.001 percent of the weight of
acetone used be subtracted from the sample
weight.
  3.3  Analysis. Two reagents-are required
for the analysis:
  3.3.1 Acetone. Same as 3.2.
  3.3.2  Desiccant. Anhydrous calcium sul-
fate,  indicating type. Alternatively,  other
types of desiccants may be used, subject to
the approval of the Administrator.
4. Procedure
  4.1  Sampling.  The complexity  of this
method is such that, in order to obtain reli-
able results, testers should be trained and
experienced with the test procedures.
  4.1.1  Pretest Preparation. All the compo-
nents shall be maintained and calibrated ac-
cording  to the  procedure described  in
APTD-0576,  unless   otherwise  specified
herein.
  Weigh several 200 to 300 g portions  of
silica gel in air-tight containers to the near-
est 0.5 g. Record the total weight of the
silica gel plus container, on each container.
As an alternative,  the silica gel need not be
preweighed, but may be weighed directly in
the impinger or sampling holder just prior
to train assembly.
  Check filters visually against light  for ir-
regularities and  flaws  or pinhole  leaks.
Label filters of the proper diameter on the
back  side near  the  edge using numbering
machine ink.  As  an  alternative, label the
shipping containers (glass or plastic petri
dishes) and keep the filters in these contain-
ers at all times except during sampling and
weighing.
  Desiccate the filters at 20±5.6° C <68±10'
F)  and-ambient pressure for at  least  24
hours and weigh at intervals of at least 6
hours to a constant weight,  i.e., 0.5  mg
change from previous weighing; record re-
sults  to the nearest 0.1  mg. During each
weighing the filter must not be exposed to
the  laboratory  atmosphere for a period
greater than 2 minutes and a relative hu-
midity  above  50 percent.  Alternatively
(unless otherwise specified by the Adminis-
trator), the filters  may be oven dried at 105*
C (220° F) for 2 to 3 hours, desiccated for 2
hours, and weighed. Procedures other than
those described, which  account  for relative
humidity effects,  may  be used, subject to
the approval of the Administrator.
  4.1.2  Preliminary  Determinations.  Select
the sampling site and the minimum number
of sampling points according to Method 1 or
as specified by the  Administrator. Deter-
mine the stack pressure, temperature, and
the range of velocity heads using Method 2;
it is recommended that a leak-check of the
pitot lines (see  Method 2,  Section 3.1)  be
performed. Determine the moisture content
using Approximation Method 4 or its alter-
natives for the purpose of making isokinetic
•sampling rate settings. Determine the stack
gas dry  molecular weight, as described in
Method 2, Section 3.6; if integrated Method
3 sampling is used for molecular weight de-
termination, the  integrated  bag sample
shall be taken simultaneously with, and for
                                        R-48

-------
App. A

the same total length of time as, the partic-
ulate sample run. •'
  Select a nozzle size based on the range of
velocity heads, such that it is not necessary
to change the nozzle size in order to main-
tain isokinetic sampling rates. During  the
run, do not change the nozzle size. Ensure
that the proper differental pressure gauge
is chosen for the range of velocity heads en-
countered (see Section 2.2 of Method 2).
  Select a suitable  probe liner  and probe
length such that all traverse points can be
sampled. For large stacks, consider sampling
from opposite sides of the stack to reduce
the length of probes.
  Select a total sampling time greater than
or equal  to the  minimum total sampling
time specified in the test procedures for the
specific industry such that (1) the sampling
time per point is not less  than 2 min (or
some greater time interval as specified  by
the  Administrator), and (2) the sample
volume taken (corrected  to standard condi-
tions)  will exceed the required  minimum
total gas sample volume. The latter is based
on an approximate average sampling rate.
  It  is recommended that  the number of
minutes sampled at each point be an integer
or an integer plus one-half minute, in order
to avoid timekeeping errors. The sampling
time at each point shall be the same.
  In some cirumstances, e.g., batch cycles, it
may be necessary  to sample for  shorter
times at the traverse points and to obtain
smaller gas sample volumes. In these cases,
the Administrator's approval must first be
obtained.
  4.1.3  Preparation  of  Collection  Train.
During preparation  and assembly of  the
sampling train, keep all openings where con-
tamination can  occur covered  until  just
prior to assembly or until sampling is about
to begin.
  Place 100 ml of water in each of the  first
two  impingers, leave  the  third impinger
empty, and transfer approximately 200 to
300 g of preweighed silica gel  from its  con-
tainer  to the fourth impinger. More silica
gel may be used, but care should be taken to
ensure that it is not entrained and carried
out 'from the  impinger during  sampling.
Place the container in a clean place for later
use in the sample recovery.  Alternatively,
the weight of the silica gel plus impinger
may be determined to the nearest 0.5 g and
recorded.
  Using a  tweezer or clean disposable surgi-
cal gloves, place a labeled  (identified) and
weighed filter in  the filter holder. Be  sure
that the filter is property centered and the
gasket properly placed so as to prevent the
sample gas stream from  circumventing the
filter. Check the filter for tears after assem-
bly is completed.
  When glass liners are used, install the se-
lected  nozzle using a Viton A O-ring when
stack temperatures are  less than 260° C
    Title 40—Protection of Environment

 (500° P) and an asbestos string gasket when
 temperatures are  higher. See  APTD-0576
 for details. Other connecting systems using
 either 316 stainless steel or Teflon ferrules
 may be used. When metal liners are used,
 install the nozzle as above or by a leak-free
 direct  mechanical  connection.  Mark the
 probe with heat resistant tape or by some
 other method to denote the proper distance
 into the stack or duct for each  sampling
 point.
   Set up the train as in Figure 5-1, using (if
 necessary)  a very light  coat of silicone
 grease  on all ground glass  joints, greasing
 only the outer portion (see  APTD-0576) to
 avoid possibility of contamination by the
 silicone grease. Subject to the approval of
 the Administrator,  a glass cyclone may be
 used between  the probe and filter  holder
 when the total participate catch is expected
 to exceed 100 mg or when water droplets are
 present in the stack gas.
   Place crushed ice around the impingers.
   4.1.4  Leak-Check Procedures.
   4.1.4.1  Pretest   Leak-Check.   A pretest
 leak-check  is  recommended,  but not  re-
 quired. If the tester opts to conduct the pre-
 test leak-check, the following procedure
 shall be used.
   After the sampling train has been assem-
 bled, turn on and set the filter and probe
 heating systems at the  desired  operating
 temperatures. Allow time for the  tempera-
 tures to stabilize.  If a Viton A  O-ring or
 other leak-free connection is used in assem-
 bling the probe nozzle to the probe liner,''
 leak-check the train at the sampling site by
 plugging the nozzle and  pulling a 380 mm
 Hg (15 in. Hg) vacuum.

   NOTE: A lower vacuum  may be used, pro-
 vided that it Is not exceeded during the test.
   If an asbestos string is used, do not con-
 nect the probe to the train during the leak-
 check. Instead, leak-check the train by first
 plugging the  inlet to  the filter  holder
 (cycone, if applicable) and pulling a 380 mm
 Hg (15 in. Hg) vacuum (see Note immediate-
 ly above). Then connect the probe  to the
 train and leak-check at about 25 mm Hg (1
 in. Hg) vacuum; alternatively,  the probe
 may be leak-checked with the rest of the
 sampling train, in  one step, at 380 mm Hg
 (15 in. Hg) vacuum. Leakage rates in excess
 of 4 percent of the average sampling rate or
'•"0.00057 m'/min (0.02 cfm), whichever is less,
 are unacceptable.
   The following leak-check instructions for
 the sampling train described in APTD-0576
 and APTD-0581 may be  helpful.'Start the
 pump with bypass  valve fully open and
 coarse adjust valve, completely closed. Par-
 tially open the coarse  adjust  valve  and
 slowly close the bypass valve until the de-
 sired vacuum is reached.  Do not reverse di-
 rection of bypass value; this will cause water
                                       B-49

-------
  Chapter I—Environmental Protection Agency
                                 APP.A
  to back up into the filter holder. If the de-
  sired vacuum is exceeded, either leak-check
  »t this higher vacuum or end the leak-check
  as shown below and start over.
   When the leak-check is completed, first
  slowly remove the plug from the inlet to the
  probe, filter holder, or cyclone (if applica-
  ble) and immediately turn off the vacuum
  pump. This prevents the water in the im-
  plngers from being forced backward into the
  filter holder and silica get from being en-
  trained backward into the third impinger.
   4.1.4.2 Leak-Checks During Sample Run.
 If, during  the sampling run, a component
 (e.g., filter assembly or Impinger) change be-
 comes unecessary, a leak-check shall be con-
 ducted immediately before the change is
 made. The leak-check shall be done accord-
 ing to  the procedure outlined in Section
 4.1.4.1 above, except that it shall be done at
 a vacuum equal to or greater than the maxi-
 mum value recorded up to that point in the
 test. If the leakage rate is found  to be no
 greater than 0.00057 mVmin (0.02 cfm) or 4
 percent  of  the  average sampling  rate
 (whichever is, less), the results are accept-
 able, and no correction will need to be ap-
 plied to the total volume of dry gas metered;
 if,  however,  a higher leakage rate is  ob-
 tained,  the tester shall  either record the
 leakage rate and plan to correct the sample
 volume as shown in Section  6.3  of  this
 method, or shall void the sampling run.
  Immediately  after  component  changes,
 leak-checks are optional; If such leak-checks
 are done, the procedure outlined in Section
 4.1,4.1 above shall be used.
  4.1.4.3  Post-test  Leak-Check.  A  leak-
 check  is mandatory at the  conclusion of
 each sampling run.  The leakcheck shall be
 done in accordance with the procedures out-
 lined In Section 4.1.4.1, except that it shall
 be conducted at a vacuum equal to or great-
 er than the maximum value reached during
 the sampling run.  If the  leakage rate  is
 found to be no greater than 0.00057 mVmin
 (0.02 cfm) or 4 percent of the average sam-
pling rate (whichever is less), the results are
acceptable, and no  correction need be ap-
plied to the total volume of dry gas metered.
If,  however,  a higher leakage rate is ob-
tained, the tester shall either record the
leakage rate and correct the sample volume
 as shown in Section 6.3 of this method, or
 shall void the sampling run.
   4.1.5  Participate    Train    Operation.
 During the sampling run, maintain an iso-
 kinetic sampling rate (within 10 percent of
 true isokinetic unless otherwise specified by
 the   Administrator)  and  a temperature
 around the filter of 120±14°  C (248±25° P),
 or such other temperature  as specified  by
 an applicable subpart of the standards or
 approved by the Administrator.
   For each run, record the data required on
 a data sheet such as the   one shown  in
 Figure 5-2. Be sure to record the initial dry
 gas meter reading. Record the dry gas meter
 readings at the  beginning and end of each
 sampling time increment, when changes in
 flow rates are made, before and after each
 leak-check,  and when sampling is halted.
 Take other readings required by Figure 5-2
 at least once at each sample point during
 each  time increment and additional  read-
 ings when significant  changes  (20 percent
 variation in velocity head readings) necessi-
 tate additional  adjustments in flow rate.
 Level and zero the manometer. Because the
 manometer  level and zero may drift due to
 vibrations and temperature  changes, make
 periodic checks during the traverse.
  Clean the portholes prior to the test run
 to minimize the chance of sampling deposit-
 ed material. To begin sampling, remove the
 nozzle cap, verify that the filter and probe
 heating systems  are up to temperature, and
 that the pitot tube and probe are properly
 positioned. Position the nozzle  at the  first
 traverse point with the tip pointing directly
 into the gas stream. Immediately start the
 pump and adjust the flow to  isokinetic con-
 ditions. Nomographs  are available, which
 aid in the rapid adjustment of the isokinetic
 sampling rate without  excessive computa-
 tions. These nomographs are designed for
 use when the Type S pitot tube coefficient
 is 0.85+0.02, and the stack  gas  equivalent
 density (dry molecular weight) is equal to
 29db4. APTD-0576 details the procedure for
using the nomographs. If CP and Ma are out-
side the above stated ranges do not use the
nomographs unless appropriate  steps  (see
Citation 7 in Section 7) are, taken to  com-
pensate for the deviations.
                                        B-50

-------
                                         PLANT
                                         LOCATION	
                                         OPERATOR,	
                                         DATE	
                                         RUN NO	
                                         SAMPLE BOX NO..
                                         METER BOX NO._
                                         METERAH®	
                                         C FACTOR	
AMBIENT TEMPERATURE.
BAROMETRIC PRESSURE-
ASSUMED MOISTURE. X_
PROBE LENGTH,™ (ft)	
                                         PITOT TUBE COEFFICIENT, Cp.
                                                                                 SCHEMATIC OF STACK CROSS SECTION
NOZZLE IDENTIFICATION NO	
AVERAGE CALIBRATED NOZZLE DIAMETER, cm (in.).
PROBE HEATER SFTTIMr.	
LEAK RATE,m3/mii>.(cfm)	
PROBE LINER MATERIAL	
STATIC PRESSURE, mm Hg (in. Hg)_
FILTER NO	
CO
 I
en
TRAVERSE POINT
. NUMBER












TOTAL
AVERAGE
SAMPLING
TIME
(0), min.














VACUUM
mm Hg
(in. Hg)














STACK
TEMPERATURE

-------
Chapter I—Environmental Protection Agency
                                 App.A
  When the stack is under significant nega-
tive pressure  (height  of impinger  stem),
take care  to close the coarse adjust valve
before inserting the probe into the stack to
prevent water  from backing into the filter
holder.  If  necessary,  the pump  may be
turned  on with  the coarse adjust valve
closed.
  When the probe is in position, block off
the openings around the probe and porthole
to prevent unrepresentative dilution of the
gas stream.
  Traverse the stack cross-section,  as re-
quired by  Method 1  or as specified by the
Administrator, being careful not to bump
the probe  nozzle into the stack walls when
sampling near the walls or when removing
or inserting the  probe through the port-
holes; this minimizes the chance of extract-
ing deposited material.
  During the test run, make periodic adjust-
ments to keep the temperature around the
filter holder at the proper level; add more
ice and, if  necessary, salt to maintain a  tem-
perature of less than 20' C (68° F)  at the
condenser/silica gel outlet. Also, periodical-
ly check the level and zero of the mano-
meter.
  If the pressure drop  across the filter be-
comes too high, making isokinetic sampling
difficult to maintain, the filter may be re-
placed  in  the  midst  of  a sample run.  It is
recommended  that another complete filter
assembly be used rather than attempting to
change the filter itself. Before  a new filter
assembly is installed, conduct a leak-check
(see Section 4.1.4.2). The total  particulate
weight shall Include the summation of all
filter assembly catches.
  A single train shall be used for the entire
sample run, except in cases where simulta-
neous sampling is required in two or more
separate ducts or at two or more different
locations within the  same duct, or, in cases
where   equipment  failure  necessitates  a
change of trains. In all other situations, the
use of two or more trains will be subject to
the approval of the Administrator.
  Note that when two  or more  trains are
used,  separate analyses of the  front-half
and (If applicable) impinger catches  from
each train shall be performed, unless identi-
cal nozzle sizes were used on all trains, in
which case, the front-half catches from the
individual trains may be combined (as may
the impinger  catches)  and one analysis of
front-half  catch  and one analysis  of im-
pinger catch  may be  performed.  Consult
with the Administrator for details concern-
ing the calculation of  results when two or
more trains are used.
  At the end of the sample run, turn off the
coarse adjust  valve,  remove the probe and
nozzle from the stack,  turn off the pump,
record the final dry  gas meter reading, arid
conduct a post-test leak-check,  as outlined
In Section 4.1.4.3. Also, leak-check the  pitot
lines as described in Method 2, Section 3.1;
the lines must pass this leak-check, in order
to validate the velocity head data.
  4.1.6  Calculation  of Percent Isokinetic.
Calculate percent isokinetic (see Calcula-
tions, Section 6) to determine whether the
run was valid or another test run should be
made. If there was difficulty in maintaining
isokinetic rates due  to  source conditions,
consult with the Administrator for possible
variance on the isokinetic rates.
  4.2  Sample  Recovery.  Proper cleanup
procedure begins as soon as the probe is re-
moved from the stack at the end of the sam-
pling period. Allow the probe to cool.
  When the probe can be  safely handled,
wipe off all external particulate matter near
the tip of the probe nozzle and place a cap
over it to prevent losing  or gaining particu-
late matter.  Do not cap off the probe  tip
tightly while the sampling train is cooling
down as this would create  a vacuum in the
filter holder, thus drawing water from the
impingers into the filter holder.
  Before moving the  sample train  to the
cleanup site, remove  the  probe from the
sample train, wipe off the silicone  grease,
and cap the open outlet of the probe. Be
careful not  to lose any  condensate  that
might  be present.  Wipe  off  the silicone
grease from the filter inlet where the probe
was fastened and cap it. Remove the  umbili-
cal cord from the last impinger and cap the
impinger. If a  flexible line is used between
the  first impinger  or condenser and the
filter  holder,  disconnect  the  line  at the
filter holder and let any  condensed water or
liquid drain into the impingers or condens-
er.  After wiping off the  silicone grease, cap
off the filter  holder  outlet and impinger
inlet. Either ground-glass  stoppers, plastic
caps, or serum caps may  be used to  close
these openings.
  Transfer the probe and filter-impinger as-
sembly  to  the cleanup  area.  This  area
should be  clean  and protected  from the
wind so that the chances of contaminating
or losing the sample will  be minimized.
  Save a portion  of the acetone used  for
cleanup as a blank.  Take 200 ml of this ac-
etone directly  from the wash  bottle being
used and place it in a glass sample container
labeled "acetone blank."
  Inspect the train > prior to and during dis-
assembly and note any abnormal conditions.
Treat the samples as follows:
  Container No. 1. Carefully  remove the
filter from  the filter holder and place it in
Its  identified petri dish container. Use a pair
of tweezers and/or clean disposable surgical
gloves to handle the filter. If it is necessary
to fold the filter, do so such that the partic-
ulate cake is inside the fold. Carefully trans-
fer to the petri dish any particulate matter
and/or  filter  fibers which adhere  to  the
filter  holder gasket, by  using  a dry Nylon
                                         B-52

-------
App. A

bristle brush and/or a sharp-edged blade.
Seal the container.
  Container No. 2. Taking care to see that
dust on the outside of the probe or other ex-
terior surfaces does not get into the sample,
quantitatively recover particulate matter or
any  condensate  from  the  probe  nozzle,
probe fitting, probe liner, and front half of
the filter holder by washing these compo-
nents with acetone and placing the wash in
a glass container. Distilled water  may  be
used instead of acetone  when approved  by
the Administrator and shall be used when
specified  by the Administrator;  in these
cases, save a water blank and follow the Ad-
ministrator's directions on analysis. Perform
the acetone rinses as follows: ,
  Carefully  remove  the  probe nozzle and
clean the inside surface by rinsing with ac-
etone from a wash bottle and brushing with
a Nylon bristle brush. Brush until  the ac-
etone rinse  shows no visible particles, after
which make a final rinse of the inside sur-
face with acetone.
  Brush and rinse the inside parts of the
Swagelok fitting with acetone in a similar
way until no visible particles remain.
  Rinse the probe liner with acetone by tilt-
Ing and rotating the probe while squirting
acetone into its upper end so that all inside
surfaces will be wetted with acetone. Let the
acetone drain from the  lower end into the
sample container. A  funnel (glass  or poly-
ethylene) may be used to aid on transfer-
ring liquid washes to the container. Follow
the acetone rinse with a probe brush. Hold
the probe in an inclined position, squirt ac-
etone into the upper end as the probe brush
is being pushed with  a  twisting  action
through the probe; hold a sample container
underneath the lower end of the probe, and
catch  any acetone and  particulate matter
which is brushed from the probe. Bun the
brush through the  probe three times or
more until  no visible particulate matter is
carried out with the acetone or  until none
remains in the probe liner on visual inspec-
tion. With stainless steel or other metal
probes, run the brush through in the above
prescribed manner at least six times since
metal probes have small crevices in which
particulate matter can be entrapped. Rinse
the brush with acetone, and quantitatively
collect these washings in the sample con-
tainer. After the brushing, make a final ac-
etone rinse of the probe as described above.
  It is  recommended that two people be
used to clean the probe  to minimize sample
losses. Between sampling runs, keep brushes
clean and protected from contaminations.
   Title 40—Protection of Environment

  After ensuring that all  joints have been
wiped  clean  of  silicone grease, clean the
inside of the front half of the filter holder
by rubbing the surfaces with a Nylon bristle
brush and rinsing with acetone. Rinse each
surface three times or more if needed  to
remove visible partieulate. Make  a final
rinse of the brush and filter holder. Careful-
ly rinse out the glass cyclone, also (if appli-
cable). After all acetone washings and par-
ticulate matter have been collected in the
sample container, tighten the lid  on the
sample container so that acetone will not
leak out when it is shipped to the  labora-
tory. Mark the height of the fluid levcel to
determine whether or not leakage  occured
during  transport.  Label the  container  to
clearly identify its contents.
  Container No. 3. Note the color of the  in-
dicating silica gel to determine if it has been
completely spent and make a notation of its
condition. Transfer the silica gel from the
fourth  impinger to its original container
and seal. A  funnel may make it easier to
pour the silica gel without spilling. A rubber
policeman may be used as an aid in remov-
ing the silica gel from the impinger. it is not
necessary to  remove the small amount of
dust particles that may adhere to the im-
pinger  wall  and are  difficult to  remove.
Since the gain in weight is to be used  for
moisture calculations, do not use any water
or other liquids to transfer the silica gel. If
a balance is available in the field, follow the
procedure for container No. 3 in Section 4.3.
  Impinger  Water.  Treat the impingers as
follows; Make a notation of any  color or
film in the liquid catch. Measure the liquid
which  is in  the first  three  impingers to
within ±1 ml by,using a graduated cylinder
or by weighing it to within ±0.5 g by using a
balance (if  one is available). Record the
volume or weight of liquid present. This in-
formation is required to calculate the mois-
ture content of the effluent gas.
  Discard the liquid after measuring and re-
cording the volume or weight, unless analy-
sis  of the impinger catch is required (see
Note, Section 2.1.7).
   If a different type of condenser  is used,
measure the amount of moisture condensed
either volumetrically or gravimetrically.
   Whenever possible,  containers should be
shipped in such a way that they remain up-
 right at all times.
   4.3  Analysis. Record the data required on
 a sheet such as the one shown in Figure  5-3.
 Handle each sample container as follows:
                                          B-53

-------
 Chapter I—Environmental Protection Agency
   Plant	
                             App. A
   Date.
   Run No..
  Filter No..
  Amount liquid lost during transport
  Acetone blank volume, ml	
  Acetone wash volume, ml	
  Acetone blank concentration, mg/mg (equation 54).
  Acetone wash blanx, mg (equation 5-5)	
CONTAINER
NUMBER
1
2
TOTAL
WEIGHT OF PARTICULATE COLLECTED,
mg
FINAL WEIGHT


^>-NCREASE' 9 - VOLUME WAT€R. ml
   1 g/ml
                               B-54

-------
App. A
   Title 40—Protection of Environment
  Container No. 1. Leave the contents in the
shipping container or transfer the filter and
any loose  particulate from the sample con-
tainer to a tared glass weighing dish. Desk*_
cate for 24 hours in a desiccator containing
anhydrous calcium sulfate. Weigh to a con-
stant weight and report the results to the
nearest 0.1 mg. For purposes of this Section,
4.3, the term "constant weight" means a dif-
ference of no more than 0.5 mg or 1 percent
of total weight less tare weight, whichever is
greater, between two consecutive weighings,
with no less  than  6 hours  of desiccation
time between weighings.
  Alternatively,  the sample  may  be  oven
dried at 105°  C  (220° P) for 2 to 3 hours,
cooled  in  the desiccator, and weighed to a
constant weight,  unless otherwise specified
by the Administrator.  The tester may  also
opt to  oven dry  the sample at 105° C (220°
P) for  2 to 3 hours, weigh the sample, and
use this weight as a final weight.
  Container No. 2. Note the level of liquid in
the container and confirm on  the analysis
sheet  whether  or  not leakage  occurred
during transport. If a noticeable amount of
leakage has occurred, either void the sample
or use  methods, subject to the approval of
the Administrator,  to  correct the final re-
sults. Measure the  liquid in this container
either  volumetrically to ±1 ml or gravime-
trically to ±0.5 g. Transfer the contents to a
tared 250-ml beaker and evaporate  to dry-
ness at ambient temperature and pressure.
Desiccate  for  24 hours and weigh to a con-
stant weight. Report the results to the near-
est 0.1  mg.
  Container No.  3. Weigh the spent silica
gel (or silica gel  plus impinger) to the near-
est 0.5 g using a  balance. This step may be
conducted in the field.
  "Acetone Blank" Container.  Measure ac-
etone in this container either volumentrical-
ly or gravimetrically. Transfer the acetone
to a tared 250-ml beaker and evaporate to
dryness at ambient temperature and  pres-
sure. Desiccate for 24 hours and weigh to a
constant weight. Report the results to the
nearest 0.1 mg.
  NOTE: At the option of the tester, the con-
tents of Container  No. 2 as well  as  the ac-
etone blank container may be evaporated at
temperatures higher than ambient. If evap-
oration is done at an elevated temperature,
the temperature  must be below the boiling
point of the solvent; also, to prevent  "bump-
ing," the evaporation process must be close-
ly  supervised, and the  contents  of  the
beaker must be swirled occasionally to
maintain an even temperature. Use extreme
care, as  acetone is highly flammable and
has a low  flash point.
5. Calibration
  Maintain a laboratory log of all  calibra-
tions.
  5.1  Probe Nozzle. Probe nozzles shall be
calibrated before  their initial use in the
field.  Using  a  micrometer, measure the
inside diameter of the nozzle to the nearest
01.025 mm (0.001 in.). Make three separate
measurements  using  different  diameters
each time, and  obtain the average of the
measurements. The difference between the
high and low numbers shall not exceed 0.1
mm (6.004 in.).  When   nozzles  become
nicked,  dented,  or corroded, they shall be
reshaped,  sharpened,   and   recalibrated
before use. Each nozzle shall be permanent-
ly and uniquely identified.
  5.2  Pitot Tube. The Type S pitot tube as-
sembly shall be  calibrated according to the
procedure outlined in Section 4 of Method
2.
  5.3  Metering  System.  Before  its initial
use in the field, the metering system shall
be  calibrated  according to the  procedure
outlined in APTD-0576. Instead of physical-
ly adjusting the dry gas meter dial readings
to correspond to the wet test meter read-
ings,  calibration factors  may  be used  to
mathematically  correct the gas meter dial
readings to the  proper values. Before cali-
brating  the metering system, it is suggested
that a leak-check  be conducted. For meter-
ing systems having diaphragm pumps, the
normal leak-check procedure will not detect
leakages within the pump, for these cases
the following  leak-check  procedure is sug-
gested: make a 10-minute  calibration run at
0.0057 m Vmin (0.02 cfm); at the end of the
run, take the difference  of the measured
wet test meter and dry gas meter volumes;
divide the difference by 10, to get the leak,
rate.  The leak rate should  not exceed
0.00057 m Vmin  (0.02 cfm).
  After each field use, the calibration of the
metering system shall be checked by per-
forming three calibration runs at a single,
intermediate orifice setting (based on the
previous field  test). With  the vacuum set at
the maximum value reached during the test
series. To adjust the vacuum, insert a valve
between the wet test meter and the inlet of
the metering system. Calculate the average
value of the calibration factor. If the cali-
bration has changed by  more than 5 per-
cent,  recalibrate  the  meter over the full
range  of orifice  settings, as  outlined  in
APTD-0576.
  Alternative procedures,  e.g., using the ori-
fice meter coefficients, maybe used, subject
to the approval of the Administrator.
  NOTE:  If the  dry gas  meter  coefficient
values  obtained before  and after  a test
series differ by more than 5 percent, the
test series shall  either be  voided, or calcula-
tions for test series shall be performed using
whichever meter  coefficient  value  (i.e.,
before or  after) gives the lower value of
total sample volume.
                                       B-55

-------
 Chapter I—Environmental Protection Agency
                                 App.A
  6.4 Probe Heater Calibration. The probe
 heating system shall be calibrated before its
 initial use in the field according to the pro-
 cedure outlined in APTD-0576. Probes con-
 structed according to  APTD-0581 need not
 be  calibrated  if the  calibration curves in
 APTD-0576 are used.
  5.5  Temperature Gauges. Use the proce-
 dure in Section 4.3 of Method 2 to calibrate
 In-stack temperature gauges. Dial thermom-
 eters, such as are used for the dry gas meter
 «nd condenser outlet, shall be calibrated
 against mercury-In-glass thermometers.
  5.6  Leak  Check of  Metering System
 Shown in Figure 5-1. That portion of the
 sampling train from the pump to the orifice
 meter should be leak checked prior to initial
 use and  after each shipment. Leakage after
 the pump will result in less volume being re-
corded than is actually sampled. The follow-
ing  procedure is suggested (see Figure 5-4):
 Close  the main  valve  on the meter box.
 Insert  a  one-hole  rubber  stopper  with
 rubber tubing atached  into the orifice ex-
 haust pipe. Disconnect and vent the low side
 of the orifice manometer. Close off the low
 side orifice tap. Pressurize the system to  13
 to 18 cm (5 to 7 in.) water column by blow-
 ing into the rubber tubing. Pinch'off the
 tubing and observe the  manometer for one
 minute. A loss of pressure on the  mano-
 meter  indicates  a leak  in the meter  box;
 leaks, if present, must be corrected.
  5.7  Barometer. Calibrate against a mer-
 cury barometer.
 6. Calculations
  Carry out calculations, retaining at least
 one extra decimal figure beyond that of the
 acquired data. Round off figures  after the
 final calculation. Other  forms of the equa-
tions may be used as long as they give equiv-
alent results.
                                      B.-56

-------
tn
                                                                                            VACUUM
                                                                                             GAUGE
                                  RUBBER
                                  TUBING
                      BLOW INTO TUBING
                      UNTIL MANOMETER
                     READS 5 TO 7 INCHES
                       WATER COLUMN      ORIFICE
                                        MANOMETER
AIR-TIGHT
  PUMP
I
a
o
                                                          figure 5-4. Leak check of meter box.

-------
 Chapter I—Environmental Protection Agency
                                 APP.A
   6.1 Nomenclature

 X.=Cross-sectional area of nozzle, m2 (ft2).
 BM=Water vapor in the gas stream, propor-
    tion by volume.
 G=Acetone  blank residue concentration,
    mg/g.
 c,=Concent rat Ion .of participate matter in
    stack gas, dry basis, corrected to stand-
    ard conditions, g/dscm (g/dscf).
 7=Percent of isokinetic sampling.
 Zo=Maximum acceptable leakage  rate for
    either a pretest leak check or for a leak
    check following a component change;
    equal to  0.0057  m'/min (0.02 cfm) or 4
    percent of the  average sampling rate,
    whichever is less.
 l4=Individual leakage rate observed during
    the leak  check  conducted  prior to the
    "i"1" component change (4=1, 2, Z....n),
    m'/mln (cfm).
 Ip=Leakage rate observed during the post-
    test leak check, mVmin (cfm).
 m.=Total amount of particulate matter col-
    lected, mg.
 if»=Molecular weight of water,  18.0 g/g-
    mole (18.01b/lb-mole).
 ma=Mass of residue of acteone after evapo-
    ration, mg.
 P^=Barometric pressure at the sampling
   site, mm Hg (in.  Hg).
 P,=Absolute stack gas pressure, mm Hg (in.
   Hg).
 P,u=Standard absolute pressure, 760 mm
   Hg (29.92  in. Hg).
 H=Ideal gas  constant, 0.06236 mm Hg-m3/
    •K-g-mole (21.85 in. Hg-ftVR-lb-mole).
 Tm=Absolute average dry gas  meter tem-
   perature (see Figure 5-2), °K (°B).
 T,—Absolute average stack gas temperature
   (see Figure 5-2), °K (°R).
 71IU,=Standard absolute temperature, 293° K
   (528' R).
 V,=Volume of acetone blank, ml.
 V«=Volume of acetone used in wash, ml.
 Vif=Total volume of liquid collected in im-
   pingers and silica gel (see Figure 5-3),
   ml.
 V»,= Volume of gas  sample as measured by
   dry gas meter, dcm (dscf).
 Vn (,,,<)=Volume of gas sample measured by
   the dry gas meter, corrected to standard
   conditions, dscm (dscf).
 lk<«
-------
 App. A

 and substitute only for those leakage rates
 (Li or Lp) which exceed La.
   6.4  Volume of water vapor.

                                Equation 5-2
                       Title 40—Protection of Environment

                      6.7 Acetone Wash Blank.


                                 »F. = C.V..p.
                                                   Equation 5-5
 where:
 #,=0.001333 m'/ml for metric units
   =0.04707 ftVml for English units.
 6.5 Moisture Content.
           Ba.,-
                                Equation 5-3
  NOTE: In saturated or water droplet-laden
 gas streams, two  calculations of the mois-
 ture content of the stack gas shall be' made,
 one from the impinger analysis (Equation 5-
 3), and a  second from  the assumption of
 saturated conditions. The lower of the two
 values of Ba shall be considered correct. The
 procedure  for determining the  moisture
 content based upon assumption of saturated
 conditions  is given in the Note of Section 1.2
 of  Method  4. For  the purposes  of this
 method, the average stack gas temperature
 from Figure 5-2 may be used to make this
 determination, provided that the accuracy
 of the in-stack temperature sensor is ±1° C
 (2° F).

  6.6 Acetone Blank Concentration.
                      6.8  Total Particulate Weight. Determine
                    the total partieulate catch from the sum of
                    the weights obtained from containers 1 and
                    2 less the acetone blank (see Figure 5-3).

                      NOTE: Refer to Section 4.1.5 to assist in
                    calculation of results involving two or more
                    filter assemblies or two  or more sampling
                    trains.
                      6.9 Particulate Concentration.
                          c.= (0.001 ff/mg)
                     6.10 Conversion Factors:
                                                   Equation 5-6
From
scf 	
g/ft' 	
g/ft3
g/fts. 	

To
m*
gr/fl5. 	
lb/fts


Multiply by
0 02832
15.43.
2 205X10"3
35 31

                               Equation 5-4
                     6.11 Isokinetic Varition.
                     6.11.1 Calculation From Raw Data.
100 T,[K3Vle
                                       606v.P,An
                                              ) ( Pbtr+ Aff/13.6)]
                                                  Equation 5-7
where:
#,=0.003454 mm Hg-m3/ml-°K for metric
    units.
   =0.002669-in. Hg-ft'/ml-'R for English
    units.
  6.11.2  Calculation  From   Intermediate
Values.
          --K4
                  T V
                  * »vi
                     m («td)
                               Equation 5-8
                   where:
                   ,ff.=4.320 for metric units
                    =0.09450 for English units.
                     6.12  Acceptable Results. If 90 percent /
                    <110< percent, the results are acceptable.
                   If the results  are low in comparison to the
                   standard and / is  beyond  the acceptable
                   range, or, if / is less than 90 percent, the Ad-
                   ministrator may opt to accept the results.
                                         B-59

-------
Use Citation 4 to make judgments. Other-
wise, reject the results and repeat the test.
7. Bibliography
  1. Addendum to Specifications for Inciner-
ator Testing  at Federal Facilities.  PHS,
NCAPC. Dec. 6, 1967.
  2. Martin, Robert M. Construction Details
of Isokinetic  Source-Sampling Equipment.
Environmental  Protection   Agency.  Re-
search  Triangle Park,  N.C. APTD-0581.
April 1971.
  3. Rom, Jerome J. Maintenance, Calibra-
tion, and  Operation of Isokinetic Source
Sampling Equipment. Environmental Pro-
tection Agency.  Research Triangle  Park,
N.C. APTD-0576. March, 1972.
 4. Smith, W. S., R. T. Shigehara, and W.
P. Todd. A method of Interpreting  Stack
Sampling Data. Paper Presented at the 63d
Annual Meeting  of the Air Pollution Con-
trol Association, St. Louis, Mo. June  14-19,
1970.
 5. Smith, W. S., et al. Stack Gas Sampling
Improved and Simplified With New Equip-
ment. APCA Paper No. 67-119.1967.
 6. Specifications for Incinerator Testing at
Federal Facilities. PHS, NCAPC. 1967.
 7. Shigehara,  R. T. Adjustments in the
EPA Nomograph for Different Pitot Tube
Coefficients  and Dry Molecular  Weights.
Stack Sampling News 2:4-11, October, 1974.
 8. Vollaro, R. F. A Survey of Commercially
Available Instrumentation For the Measure-
ment of Low-Range Gas Velocities. U.S. En-
vironmental Protection Agency,  Emission
Measurement Branch.  Research  Triangle
Park,  N.C. November, 1976  (unpublished
paper).
  9. Annual Book of ASTM Standards. Part
26.  Gaseous Fuels; Coal and Coke; Atmos-
pheric Analysis. American Society for Test-
ing and Materials. Philadelphia, Pa.  1974.
pp. 617-622.
  10. Felix, L. G., G. I. Clinard, G. E. Lacey,
and J. D. McCain. Inertial Cascade Impac-
tor Substrate Media for Flue Gas Sampling.
U.S. Environmental Protection Agency. Re-
search  Triangle Park, N.C. 27711, Publica-
tion No. EPA-600/7-77-060. June 1977. 83 p.
                  B-60

-------
APR. A

METHOD 17—DETERMINATION OF PARTICULATE
  EMISSIONS PROM STATIONARY SOURCES (!N-
  STACK FILTRATION METHOD)

              Introduction

  Particulate matter is not  an  absolute
quantity; rather, it is a function of tempera-
ture and  pressure. Therefore, to  prevent
variability  in  particulate matter  emission
regulations and/or associated test methods,
the temperature and pressure at which par-
ticulate matter is to be measured must be
carefully defined. Of the two variables (i.e.,
temperature and pressure), temperature has
the greater effect upon  the amount of par-
ticulate matter in an effluent gas stream; in
most stationary source categories, the effect
of pressure appears to be negligible.
  In method 5, 250° F  is established as a
nominal   reference   temperature.  Thus,
where Method 5 is specified in an applicable
subpart of the standards, particulate matter
is defined with respect  to  temperature. In
order to maintain a  collection temperature
of 250° F, Method 5 employs a heated glass
sample  probe  and a heated filter holder.
This equipment is somewhat cumbersome
and requires care in its operation. There-
fore, where particulate matter concentra-
tions (over the normal range of temperature
associated with a specified source category)
are known to be independent of tempera-
ture, it is desirable  to eliminate  the glass
probe and heating systems, and sample  at
stack temperature.
  This method describes an  in-stack sam-
pling system and sampling procedures for
   Title 40—Protection of Environment

use in such  cases. It is intended to be used
only when specified by an applicable sub-
part of the  standards, and only within the
applicable temperature limits (if specified),
or when otherwise approved by the Admin-
istrator.
  1. Principle and Applicability.
  1.1  Principle. Particulate matter is with-
drawn isokinetically from  the source and
collected on a glass fiber filter maintained
at stack temperature. The particulate mass
is determined gravimetrically after removal
of uncombined water.
  1.2  Applicability. This method applies to
the determination of particulate emissions
from  stationary sources  for  determining
compliance  with new source  performance
standards, only  when specifically  provided
for in an applicable subpart of the stand-
ards. This method is not applicable to stacks
that contain liquid droplets or are saturated
with water vapor. In addition, this method
shall not be used as written if the projected
cross-sectional area of the probe extension-
filter holder assembly covers more than 5
percent of the stack cross-sectional area (see
Section 4.1.2).
  '2. Apparatus.
  2.1  Sampling Train. A schematic  of  the
sampling train used in this method is shown
in Figure  17-1. Construction  details  for
many, but not all, of the train components
are given in APTD-0581 (Citation  2 in Sec-
tion 7); for changes from  the APTD-0581'
document and  for  allowable modifications
to Figure 17-1, consult with the Administra-
tor.
                                        B-61

-------
                                                               TEMPtRATVtE     MOTUK
                                                                  IEKOR     FILTER HOLDER
                                                   «.y>1Jc»(t;iW
                                                                                                                             HWlPISERTRAtllOrriOIIAI., MAY in REPLACED
                                                                                                                                   IVAN EQUIVALENT COKDEIiSER
                                                                                                                                                                      THERMOMETER
O3
 I
                                                                                 ORIFICE MANOMETER


                                                             * SUGGESTED (INTERFERENCE-FREE) SMCINGS
DRV GAS METER
                                                                   •§
                                                                    r
                                                                    I
                                                                    i
                                                                                                                                                                                       •a
                                                                                        Figure 17-1. Paniculate-Sampling Train, Equipped with In-Stack Filter.

-------
App. A

  The  operating and  maintenance  proce-
dures for many  of the sampling train com-
ponents are described in APTD-0576 ^Cita-
tion 3  in Section 7). Since correct usage is
important  in  obtaining valid results, all
users should read the APTD-0576 document
and adopt  the operating and maintenance
procedures outlined  in it, unless otherwise
specified herein. The  sampling train  con-
sists of the following components:
  2.1.1  Probe Nozzle.  Stainless steel (316)
or glass, with sharp, tapered leading edge.
The angle of taper shall  be 30° and  the
taper shall be on the outside to preserve  a
constant  internal  diameter.  The  probe
nozzle shall be of the button-hook or elbow
design, unless otherwise specified by the Ad-
ministrator. If made of stainless steel, the
nozzle shall be  constructed  from seamless
tubing. Other materials of construction may
be used subject  to the approval of the Ad-
ministrator.
  A range  of sizes  suitable for isokinetic
sampling should be available, e.g., 0.32 to
1.27 cm (.VB to  % In)—or  larger if  higher
volume sampling trains are used—inside di-
ameter (ID) nozzles in increments of 0.16 cm
(Vie in). Each nozzle shall be calibrated ac-
cording to the procedures outlined in Sec-
tion 5.1.
  2.1.2  Filter Holder.  The  in-stack filter
holder shall be constructed of borosilicate
or quartz glass, or stainless steel; if a gasket
is used, it shall  be made of silicone rubber,
TeHon, or  stainless steel. Other holder and
gasket materials may be used subject to the
approval of the Administrator. The filter
holder shall be  designed to provide  a posi-
tive seal against leakage from the outside or
around the filter.
  2.1.3  Probe Extension. Any suitable rigid
probe extension may be used after the filter
holder.
  2.1.4 Pitot Tube. Type S, as described in
Section 2.1 of Method  2, or other device ap-
proved by  the Administrator; the pitot tube
shall be attached to the probe extension to
allow  constant monitoring of the stack gas
velocity (see Figure 17-1). The impact (high
pressure) opening plane of the pitot tube
shall be even with or above the nozzle entry
plane  during  sampling (see  Method  2,
Figure 2-6b). It is  recommended: (1)  that
the pitot tube have a known baseline coeffi-
cient, determined as outlined in Section 4 of
Method 2; and  (2) that this known coeffi-
cient be preserved by placing the pitot tube
in an interference-free arrangement with re-
spect  to the sampling nozzle, filter holder,
and temperature sensor (see Figure 17-1).
Note that  the 1.9 cm (0.75 in) free-space be-
tween the nozzle and pitot tube shown  in
Figure 17-1, is based on a 1.3 cm (0.5 in) ID
nozzle. If the sampling train is designed for
sampling at higher flow rates than that de-
 scribed in APTD-0581, thus  necessitating
 the use of larger sized nozzles, the  free-
   Title 40—Protection of Environment

space shall be 1.9 cm (0.75 in) with the larg-
est sized nozzle in place.
  Source-sampling assemblies  that do not
meet the minimum spacing requirements of
Figure 17-1  (or the equivalent of these re-
quirements,  e.g..  Figure 2-7 of Method 2)
may be used; however, the pitot tube coeffi-
cients  of  such assemblies shall be deter-
mined by calibration, using methods subject
to the approval of the Administrator.
  2.1.5  Differential  Pressure  Gauge.  In-
clined  manometer  or  equivalent  device
(two), as described in Section 2.2 of Method
2. One manometer shall be used for velocity
head (Ap) readings, and the other, for ori-
fice differential pressure readings.
  2.1.6  Condenser. It is recommended that
the impinger system described in Method  5
be used to determine the moisture content
of the stack gas. Alternatively, any  system
that allows measurement of both the water
condensed and the moisture leaving the con-
denser, each to within 1 ml or 1 g, may be
used. The moisture leaving the condenser
can  be  measured either by: (1) monitoring
the temperature and pressure at the exit of
the  condenser and  using  Dalton's  law of
partial  pressures; or (2) passing the  sample
gas  stream  through a  silica gel trap with
exit gases kept below 20° C (68° F) and de-
termining the weight gain.
   Flexible tubing may be used between the
probe extension and condenser.  If means
other than silica gel are used to determine
the  amount of moisture leaving the con-
denser, it is recommended that silica gel still
be used between the condenser system and
pump to prevent moisture condensation  in
the pump and metering devices and to avoid
the  need  to make corrections for moisture
in the metered volume.
   2.1.7  Metering  System. Vacuum gauge,
leak-free  pump, thermometers capable  of
measuring temperature to within 3° C (5.4°
F),  dry gas meter capable  of  measuring
volume to  within 2 percent, and  related
equipment,  as shown in Figure 17-1. Other
metering  systems  capable of maintaining
sampling  rates within 10 percent of isokine-
tic and of determining sample volumes  to
within  2 percent may be used, subject to the
approval  of'the Administrator.  When the
metering system is used in conjunction with
a pitot tube, the system shall enable checks
of isokinetic rates.
   Sampling  trains utilizing  metering sys-
tems designed for higher flow  rates than
that described in APTD-0581 or APTD-0576
may be used provided that  the specifica-
 tions of this method are met.
   2.1.8  Barometer.  Mercury, aneroid,   or
other barometer capable  of  measuring at-
 mospheric pressure to within 2.5 mm Hg
 (0.1 in. Hg). In many cases, the barometric
 reading may be obtained from a nearby na-
 tional weather service station, in which case
                                          B-63

-------
  Chapter I—Environmental Protection Agency
                                  App. A
  Che station value (which  is  the absolute
  barometric pressure) shall be requested and
  in adjustment for elevation differences be-
  tween  the weather  station  and sampling
  point shall be applied at a rate of minus 2.5
  mm Hg (0.1 in. Hg) per 30 m (100 ft) eleva-
  tion increase or vice  versa for elevation de-
  crease.
   2.1.9  Oas Density Determination  Equip-
 ment.  Temperature  sensor  and  pressure
 gauge, as described in Sections 2.3 and 2.4 of
 Method 2, and gas analyzer, if necessary, as
 described in Method 3.
   The temperature sensor shall be attached
 to either the pitot tube or to the probe ex-
 tension, In a fixed configuration. If the  tem-
 perature sensor is attached in the field; the
 sensor  shall be placed  in an interference-
 free, arrangement with respect to the Type
 8 pitot tube openings (as shown in  Figure
 17-1 or in Figure 2-7  of Method 2). Alterna-
 tively, the temperature  sensor need  not be
 attached  to either the  probe extension or
 pitot tube during sampling, provided that a
 difference of not more than 1  percent in the
 average velocity measurement is introduced.
 This  alternative is subject to  the approval
 of the Administrator.
  2.2  Sample Recovery.
  2.2.1  Probe Nozzle Brush.  Nylon  bristle
 brush with stainless steel wire handle.  The
 brush shall be properly sized and shaped to
 brush out the probe nozzle.
  2.2.2  Wash  Bottles—Two.   Glass  wash
 bottles   are  recommended;   polyethylene
 wash bottles may be used at  the  option of
 the tester. It is recommended that acetone
 not be stored in  polyethylene bottles for
 longer than a month.
  2.2.3  Glass Sample Storage Containers.
 Chemically resistant, borosilicate glass  bot-
 tles, for acetone washes, 500 ml or 1000 ml.
 Screw cap  liners  shall  either be rubber-
 backed  Teflon or shall be constructed so as
 to be leak-free and  resistant to chemical
 attack by acetone. (Narrow mouth glass  bot-
 tles have been found to be less  prone to
 leakage.) Alternatively, polyethylene bottles
 may be used.
  2.2.4  Petri  Dishes. For  filter samples;
 glass or polyethylene, unless otherwise spec-
 ified by the Administrator.
  2.2.5  Graduated Cylinder   and/or  Bal-
 ance.  To measure condensed water to within
 1 ml or 1  g. Graduated cylinders shall have
subdivisions no greater than 2 ml. Most lab-
 oratory balances are capable of weighing to
 the nearest 0.5 g or less. Any  of these  bal-
 ances is suitable for use here and in Section
 2.3.4.
  2.2.6  Plastic  Storage  Containers.  Air
tight containers to store silica gel.
  2.2.7  Funnel and Rubber Policeman. To
aid In transfer of silica gel to container; not
necessary if silica gel is weighed in the field.
  2.2.8  Funnel. Glass, or polyethylene, to
aid In sample recovery.
  2.3  Analysis.
  2.3.1 Glass Weighing Dishes.
  2.3.2 Desiccator.
  2.3.3 Analytical Balance. To measure to
 within 0.1 mg.
  2.3.4 Balance. To  measure to within 0.5
 mg.
  2.3.5 Beakers. 250 ml.
  2.3.6 Hygrometer.  To measure the rela-
 tive humidity of the  laboratory  environ-
 ment.
  2.3.7 Temperature Gauge.  To  measure
 the temperature of the laboratory environ-
 ment.
  3. Reagents.
  3.1  Sampling.
  3.1.1 Filters. The in-stack filters shall be
 glass mats or thimble fiber filters, without
 organic binders,  and shall exhibit at least
 99.95 percent efficiency ( 0.05 percent pene-
 tration)  on  0.3  micron dioctyl phthalate
 smoke particles.  The filter efficiency tests
 shall  be conducted in  accordance  with
 ASTM standard method D  2986-71. Test
 data from the supplier's quality control pro-
 gram are sufficient for this purpose.
  3.1.2 Silica Gel. Indicating type, 6- to 16-
 mesh. If previously used, dry at 175° C (350°
 F) for 2 hours. New silica gel may be used as
 received. Alternatively,  other types of desic-
 cants  (equivalent or better)  may be used,
 subject to the approval of the Administra-
 tor.
  3.1.3 Crushed Ice.
  3.1.4 Stopcock Grease. Acetone-insoluble,
 heat-stable silicone grease. This is  not nec-
 essary if screw-on connectors with Teflon
 sleeves, or similar, are  used.  Alternatively,
 other types of stopcock grease may be used,
 subject to the approval of the Administra-
 tor.
  3.2  Sample  Recovery.  Acetone,  reagent
 grade, 0.001 percent residue, in glass bottles.
Acetone  from metal containers generally
has a high residue blank and  should not be
used. Sometimes, suppliers transfer acetone
to  glass  bottles  from metal containers.
Thus,  acetone  blanks shall be run prior to
field use and only acetone with low blank
values ( 0.001 percent) shall be used. In no
case shall a blank value of greater than
0.001 percent of the weight of acetone used
be subtracted from the sample weight.
  3.3 Analysis.
  3.3.1 Acetone. Same as 3.2.
  3.3.2 Desiccant. Anhydrous calcium  sul-
fate,  indicating type. Alternatively, other
types of  desiccants may be used, subject to
the approval  of the Administrator.
  4. Procedure.
  4.1  Sampling.  The complexity  of this
method is such that, in  order to obtain reli-
able results,  testers should  be trained, and
experienced with the test procedures.
  4.1.1 Pretest  Preparation.   All  compo-
nents shall be maintained and calibrated ac-
                                           B-64

-------
App. A

cording  to the procedure  described  in
APTD-0576,   unless   otherwise   specified
herein.
  Weigh several 200  to  300 .g portions of
silica gel in air-tight containers to the near-
est 0.5  g. Record  the total weight of the
silica gel plus  container,  on each container.
As an alternative, the silica gel need not be
preweighed, but may  be weighed directly in
its impinger or sampling holder just prior to
train assembly.
  Check filters visually against light for ir-
regularities and flaws  or pinhole  leaks.
Label filters of the proper size on the back
side near the edge  using  numbering  ma-
chine ink. As an alternative, label the ship-
ping containers (glass or plastic petri dishes)
and keep the  filters  in these  containers at
all times except during sampling and weigh-
ing.
  Desiccate the filters at 20±5.6° C (68±10°
F)  and ambient pressure for at least 24
hours and weigh at intervals of at least 6
hours  to a constant weight, i.e., 0.5 mg
change from previous weighing;  record re-
sults to the nearest  0.1  mg. During each
weighing the filter must not be exposed to
the  laboratory  atmosphere -  for  a period
greater than 2 minutes  and a relative hu-
midity  above  50  percent.  Alternatively
(unless otherwise specified by the Adminis-
trator), the filters may be oven dried at 105°
C (220° P) for 2 to 3 hours, desiccated for 2
hours, and weighed. Procedures other than
those described, which account for relative
humidity effects, may be used, subject to
the approval of the Administrator.
   Title 40—Protection of Environment

  4.1.2 Preliminary Determinations. Select
the sampling site and the minimum number
of sampling points according to Method 1 or
as specified by the Administrator. Make a
projeeted-area model of the  probe exten-
sion-filter holder assembly, with the  pitot
tube face openings positioned along the cen-
terline of the stack, as shown in Figure 17-2.
Calculate the estimated cross-section block-
age, as shown in Figure 17-2. If the blockage
exceeds 5 percent of the duct cross sectional
area,  the tester has the following options:
(Da suitable out-of-stack filtration method
may be used instead of in-stack filtration; or
(2) a special in-stack arrangement, in which
the  sampling  and  velocity  measurement
sites are separate, may be used; for details
concerning this approach,  consult with the
Administrator (see also Citation  10 in Sec-
tion 7). Determine the stack pressure, tem-
perature, and the range of velocity heads
using Method 2;  it is recommended that a
leak-check of the pitot lines (see Method 2,
Section 3.1)  be  performed. Determine the
moisture   content  -using  Approximation
Method 4 or its alternatives for the purpose
of making  isokinetic sampling rate settings.
Determine  the  stack  gas dry  molecular
weight, as described in  Method  2, Section
3.6; if integrated Method 3 sampling is used
for molecular weight determination, the in-
tegrated bag sample shall be taken simulta-
neously with, and for the same total length
of time as,  the particular sample run.
                                           B-65

-------
   Chapter I—Environmental Protection Agency
                                   App. A
                                                           STACK
                                                           WALL
         IN-STACK FILTER-
        PROBE EXTENSION
           ASSEMBLY
                      ESTIMATED
                      BLOCKAGE
                        .(*)
PSHADED AREA]
DUCT AREAJ
X  100
Figure 17-2. Projected-area model of cross-section blockage (approximate average for
a sample traverse) caused by an in-stack filter holder-probe extension assembly.
                                      B-66

-------
 APP. A
   Title 40—Protection of Environment
  Select a nozzle size based on the range of
 velocity heads, such that it is not necessary
 to change the nozzle size in  order to main-
 tain isokinetic sampling rates. During the
 run, do not change the nozzle size.  Ensure
 that the proper differential pressure gauge
 is chosen for the range of velocity heads en-
 countered (see Section 2.2 of Method 2).
  Select a probe extension length such that
 all traverse points can be sampled. For large
 stacks,  consider  sampling from  opposite
 sides of the stack to reduce the length of
 probes.
  Select a total sampling time greater than
 or  equal to the  minimum total sampling
 time specified in the test procedures for the
 specific industry such that (1) the sampling
 time per point is not less than 2 minutes (or
 some  greater time interval if specified by
.the  Administrator), and  (2)  the  sample
 volume taken (corrected to standard condi-
 tions) will  exceed the required minimum
 total gas sample volume. The latter is based
 on an approximate average sampling rate.
  It is recommended that the number of
 minutes sampled at each point be an integer
 or an integer plus one-half minute, in order
 to avoid timekeeping'errors.
  In some circumstances, e.g., batch cycles,
 it may be necessary to sample for shorter
 times at the traverse points and to obtain
 smaller gas  sample volumes.  In these cases,
 the Administrator's approval must first be
 obtained.
  4.1.3 Preparation  of  Collection   Train.
 During  preparation  and assembly  of the
 sampling train, keep all openings where con-
 tamination  can  occur  covered until  just
 prior to assembly or until sampling is about
 to begin.
  If impingers are used to condense stack
 •gas moisture, prepare them as follows: place
 100 ml of water In each of the first two im-
 pingers, leave the  third impinger  empty,
 and transfer approximately 200 to 300 g of
 preweighed silica gel from its container to
 the fourth impinger. More silica gel may be
 used,  but care should be taken to ensure
 that it is not entrained and carried out from
 the  impinger during sampling.  Place the
 container in a clean place for later  use in
 the  sample  recovery.   Alternatively,  the
 weight of the silica gel  plus impinger may
 be determined to the nearest 0.5 g and re-
 corded.
  If some means other than impingers  is
 used to condense moisture, prepare the con-
 denser (and,  if  appropriate,  silica  gel for
 condenser outlet) for use.
  Using a tweezer or clean disposable surgi-
 cal gloves, place a labeled (identified) and
 weighed filter in the filter holder. Be sure
 that  the filter is properly centered and the
 gasket properly placed so as not to allow the
 sample gas stream to circumvent the filter.
 Check filter for tears after assembly is com-
 pleted. Mark the probe extension with heat
resistant tape or by some other method to
denote the proper distance into the stack or
duct for each sampling point.
  Assemble the train as in Figure 17-1, using
a very light coat of silicone grease on all
ground glass joints and greasing  only  the
outer portion (see APTD-0576) to avoid pos-
sibility  of contamination  by the silicone
grease.  Place  crushed ice around the  im-
pingers.
  4.1.4  Leak Check Procedures.
  4.1.4.1  Pretest Leak-Check.  A  pretest
leak-check is  recommended,  but hot  re-
quired. If the tester opts to conduct the pre-
test leak-check,  the following procedure
shall be used.
  After the sampling train has been assem-
bled, plug the inlet to the probe nozzle with
a material that will be able to withstand the
stack temperature.  Insert the. filter holder
into the stack and wait approximately 5
minutes  (or longer, if necessary)  to allow
the system to come to equilibrium with the
temperature of the stack gas stream. Turn
on the pump and draw a vacuum of at least
380 mm Hg (15 in. Hg); note that a lower
vacuum may be used, provided that it is not
exceeded during the test.  Determine  the
leakage rate. A leakage rate in excess of 4
percent  of the average sampling rate or
0.00057 m'/min. (0.02  cfm),  whichever is
less, is unacceptable.
  The following leak-check  instructions for
the sampling train described in APTD-0576
and APTD-0581 may be helpful. Start the
pump with by-pass valve fully open  and
coarse adjust valve completely closed. Par-
tially open the  coarse adjust valve  and
slowly close the by-pass valve until the de-
sired vacuum  is reached. Do not reverse di-
rection of  by-pass  valve.  If  the desired
vacuum  is exceeded, either leak-check at
this higher vacuum or end the leak-check as
shown below and start over.
  When  the leak-check  is completed, first
slowly remove the plug from the inlet to the
probe nozzle and immediately turn off the
vacuum  pump. This prevents water from
being forced backward and keeps  silica gel
from being entrained backward.
  4.1.4.2 Leak-Checks During Sample Run.
If,  during  the sampling run,  a component
(e.g., filter assembly or impinger) change be-
comes necessary, a leak-check shall be con-
ducted  immediately before the change is
made. The leak-check shall  be done accord-
ing  to  the procedure outlined in Section
4.1.4.1 above, except that it shall be done at
a vacuum equal to or greater than the maxi-
mum value recorded up to that point in the
test. If the leakage rate is  found  to be no
greater than 0.00057 mVmin (0.02  cfm) or 4
percent  of  the  average  sampling  rate
(whichever is less), the  results  are  accept-
able, and no correction will need  to be ap-
plied to the total volume of dry gas metered;
                                         B-67

-------
Chapter I—Environmental Protection Agency
                                App. A
if, however, s higher  leakage rate is ob-
tained, the tester shall either  record the
leakage rate and plan to correct the sample
volume as shown in Section 6.3 of this
method, or shall void the sampling run.
  Immediately after component  changes,
leak-checks are optional; if such leak-checks
are done, the procedure outlined in Section
4.1.4.1 above shall be used.
  4.1.4.3  Post-Test  Leak-Check.  A   leak-
check  Is mandatory at the conclusion of
each sampling run. The leak-check shall be
done in accordance with the procedures out-
lined in Section 4.1.4.1, except that it  shall
be conducted at a vacuum equal to or great-
er than the maximum value reached during
the sampling run.  If the leakage rate is
found to be no greater than 0.00057 m'/mln
(0.02 cfm) or 4 percent of the average sam-
pling rate (whichever Is less), the results are
acceptable, and no  correction need be ap-
plied to the total volume of dry gas metered.
If,  however, a higher  leakage  rate Is ob-
tained, the tester shall either  record the
leakage rate and correct the sample volume
as shown in Section 6.3 of this method, or
shall void the sampling run.
  4.1.5 Particulate    Train    Operation.
During the sampling run, maintain a sam-
pling  rate such that sampling is within 10
percent of true isokinetic, unless otherwise
specified by the Administrator.
  For each run, record the data required on
the example data sheet shown in Figure 17-
3. Be sure to record the initial dry gas meter
reading. Record the dry gas meter readings
at the beginning and end of each sampling
time increment, when changes in flow rates
are made, before and after each leak check,
and when sampling is halted. Take other
readings required by  Figure  17-3  at least
once at each sample point during each time
increment and additional readings when sig-
nificant changes (20 percent variation in ve-
locity head readings) necessitate additional
adjustments in flow rate. Level and zero the
manometer.  Because the  manometer level
and zero may drift due to vibrations and
temperature changes, make periodic checks
during the traverse.
                                        E-68

-------
App. A
Title 40—Protection of Environment






U
cc
8
ce
a.
S
£
|
















1






*
i
i-
i
s














K
r
i
u




2
E
t-
(9
Z
UJ
_J
Z
o
PROBE EXTEN
















u
»-






a
z
E
C
5
c
UL
t-
z
UJ
c
u
R
a
z
















e
x


1
CC




G
i
t
=













!'s
o **

X *"






i
.£
'E
in
I*
<
CC
X
<
UJ
«J
















H




•S
=
f
. E
• c
- uf
GC
3
STATIC PRESS
















i




•




g
i
g
u
<
g
u
E:
?
1


n.
U
u.
u.
UI
s
m

O
K '
SoSj
fi Ji
^
' i|
*i
UJ °
§5


^
ac P CA ui
fpi
0
Xffl*
S*"
• 	 •
i£
«C


•
\
I





3 til
s^
IS
.
5 =

£"?
io
jji
E?l
ilS


" E •—
s 	
e
3
™ -E
r ffS,
s °u

III


is
UJ JH

r




























































































































































































































































































































O1.
^

O)

















«







I















UJ
U
<
?
                                                                      I
                                                                      T
                                   B-69

-------
 Chapter I—Environmental Protection Agency
                                 App. A
   Clean the portholes prior to the test run
 to minimize the chance of sampling the de-
 posited material. To begin sampling, remove
 the nozzle cap and verify that the pitot tube
 and probe  extension  are properly  posi-
 tioned. Position the nozzle at the first tra-
 verse point with the tip pointing directly
 into the gas stream. Immediately start the
 pump and adjust the flow to isokinetic con-
 ditions. Nomographs are  available,  which
 aid in the rapid adjustment to the isokinetic
 sampling rate without excessive computa-
 tions. These nomographs are designed  for
 use when the Type S pitot tube coefficient
 is  0.85±0.02, and the stack gas equivalent
 density (dry molecular weight) is equal to
 28 ±4. AFTD-0576 details the  procedure for
 using the nomographs. If C, and Md are out-
 side the above stated ranges, do not use the
 nomographs unless appropriate steps (see
 Citation 7 In Section 7) are taken to com-
 pensate for the deviations.
  When the stack is under significant nega-
 tive  pressure (height  of  impinger stem),
 tajce care to close the  coarse adjust valve
 before Inserting the probe extension assem-
 bly Into the stack to prevent water from
 being forced backward. If necessary,  the
 pump may be turned on  with the coarse
 adjust valve closed.
  When the probe is in position, block off
 the openings around the probe and porthole
 to  prevent unrepresentative dilution of the
 gas stream.
  Traverse  the  stack cross section, as re-
 quired by Method 1 or as specified  by the
 Administrator,  being careful  not to  bump
 the probe nozzle into the stack  walls when
 sampling near the walls or  when removing
 or  Inserting the probe extension through
 the portholes,  to minimize chance  of ex-
 tracting deposited material.
  During the test run, take appropriate
 steps (e.g.,  adding crushed ice  to the  im-
 pinger ice bath) to maintain a temperature
 of less than 20° C (68° P) at the condenser
 outlet; this  will prevent excessive moisture
 losses. Also, periodically check the level and
 zero of the manometer.
  If the pressure drop across  the filter be-
 comes too high, making Isokinetic sampling
 difficult to  maintain, the filter may be re-
 placed in the midst of  a sample run. It is
 recommended that another complete filter
 holder  assembly be used rather than at-
 tempting to change the fliter itself. Before a
 new filter holder Is installed, conduct a leak
 check, as outlined In Section 4.1.4.2. The
 total particulate weight shall include the
 summation of all filter assembly catches.
  A single train shall be used for the entire
 sample run, except in cases where simulta-
neous sampling  is required  in two or more
separate ducts or at two or more different
locations within the same duct, or, in cases
•where  equipment  failure  necessitates  a
change of trains. In all other situations, the
 use of two or more trains will be subject to
 the approval of  the Administrator.  Note
 that when two or more trains  are used,' a
 separate  analysis of the collected particu-
 late from each train shall  be performed,
 unless identical nozzle sizes were used on all
 trains, in which case the particulate catches
 from the individual trains may be combined
 and a single analysis performed.
  At the end of the sample run, turn off the
 pump, remove the probe extension assembly
 from the stack, and record the final dry gas
 meter reading. Perform a leak-check, as out-
 lined In Section 4.1.4.3. Also, leak-check the
 pitot lines as  described in  Section 3.1 of
 Method  2; the  lines must pass this  leak-
 check, in order to validate the velocity head
 data.
  4.1.6  Calculation  of Percent Isokinetic.
 Calculate percent  isokinetic (see Section
 6.11) to determine whether another test run
 should  be made.  If  there  Is difficulty in
 maintaining  isokinetic rates due to source
 conditions, consult with the Administrator
 for possible variance on the isokinetic rates.
  4.2 Sample  Recovery.  Proper cleanup
 procedure begins as soon as the probe ex-
 tension assembly is removed  from the stack
 at the end of the sampling period. Allow the
 assembly to cool.
  When the assembly can be safely handled,
 wipe off all external particulate matter near
 the tip of the probe nozzle and  place a cap
 over it to prevent losing or gaining particu-
 late matter.  Do not cap off the probe tip
 tightly while the sampling train is cooling
 down as this would create a  vacuum in the
 filter holder, forcing condenser water back-
 ward.
  Before moving the sample train to the
 cleanup site, disconnect the filter holder-
 probe nozzle assembly from  the probe ex-
 tension; cap the open Inlet of the probe ex-
 tension. Be careful not to lose any conden-
 sate, if present. Remove the  umbilical cord
 from the condenser outlet  and  cap  the
 outlet. If a flexible line is used between the
 first Impinger (or condenser) and the probe
 extension, disconnect the line at the probe
 extension and let any condensed water or
 liquid drain into the impingers or  condens-
 er. Disconnect the probe extension from the
 condenser; cap the probe extension outlet.
After wiping  off the silicons grease, cap off
 the condenser inlet. Ground  glass stoppers,
plastic caps,  or serum caps (whichever are
 appropriate)  may  be used  to  close these
 openings.
  Transfer both the filter holder-probe
nozzle assembly and the condenser to the
cleanup area. This area should be clean and
protected from the wind so that the chances
of contaminating or losing the  sample will
be minimized.
  Save a portion of the acetone used for
cleanup as a blank. Take 200 ml of this ac-
                                         B.-70

-------
APR. A
   Title 40—Protection of Environment
etone directly from the wash bottle being
used and place it in a glass sample container
labeled "acetone blank."
  Inspect the train prior to and during dis-
assembly and note any abnormal conditions.
Treat the samples as follows:
  Container No.  1.  Carefully remove  the
filter from the filter holder and place it in
its identified petri dish container. Use a pair
of tweezers and/or clean disposable surgical
gloves to handle the filter.  If it is necessary
to fold the filter,  do so such that the partic-
ulate cake is inside the fold. Carefully trans-
fer to the petri dish any particulate matter
and/or  filter fibers which adhere  to  the
filter holder gasket, by using a dry Nylon
bristle brush and/or a sharp-edged blade.
Seal the container.
  Container No. 2.  Taking  care to see that
dust on the outside of the probe nozzle or
other exterior surfaces does not get into the
sample,   quantitatively recover  particulate
matter  or any  condensate  from the probe
nozzle, fitting,  and front half of the filter
holder by washing these components with
acetone and placing the wash in a glass con-
tainer. Distilled water, may be used instead
of acetone when  approved  by the Adminis-
trator and shall be used when specified by
the  Administrator; in these cases,  save  a
water blank and  follow Administrator's di-
rections  on  analysis. Perform the acetone
rinses as follows:
  Carefully  remove the probe nozzle  and
clean the inside surface by rinsing with ac-
etone from a wash bottle and brushing with
a Nylon bristle brush. Brush until 'acetone
rinse shows  no  visible particles, after which
make a final rinse of the inside surface with
acetone.
  Brush and rinse  with acetone the Inside
parts of the fitting in a similar way until no
visible particles remain. A  funnel (glass or
polyethylene) may be used to aid in trans-
ferring liquid washes to the container. Rinse
the brush with acetone and quantitatively
collect these washings in  the sample con-
tainer.   Between  sampling  runs,  keep
brushes clean and protected from contami-
nation.
  After  ensuring  that all  joints are wiped
clean of silicone grease (if applicable), clean
the  inside of the front half of the filter
holder by rubbing the surfaces with a Nylon
bristle  brush  and  rinsing  with acetone.
Rinse each  surface three times or more  if
needed  to remove visible particulate. Make
final rinse of the brush and filter  holder.
After all acetone washings and particulate
matter are collected in the sample contain-
er, tighten the lid on the sample container
so that  acetone will not leak out when  it  is
shipped to the  laboratory.  Mark the height
of the fluid level to determine whether or
not  leakage  occurred  during  transport.
Label the container to clearly identify  its
contents.
  Container No. 3. If silica gel is used in the
condenser system for mositure content de-
termination, note the color of the gel to de-
termine if it has been completely spent;
make a notation of its condition. Transfer
the silica gel back to its original container
and seal. A funnel may make it easier to
pour the silica gel without spilling, and a
rubber policeman may be used as an aid in
removing.the silica gel. It is not necessary to
remove the small amount of dust particles
that may adhere to the walls and are diffi-
cult to remove. Since the gain in weight is to
be used for moisture calculations,  do not use
any water  or other liquids  to  transfer the
silica gel. If a balance is available in the
field, follow the  procedure for  Container
No. 3 under "Analysis."
  Condenser  Water. Treat the condenser or
impinger water as follows: make  a notation
of any color  or film  in  the liquid catch.
Measure the liquid volume to within +1 ml
by using a graduated cylinder or, if a bal-
ance  is available,  determine the liquid
weight to within ±0.5 g. Record the total
volume or weight of liquid present. This in-
formation is required to calculate the mois-
ture content of the effluent  gas. Discard the
liquid  after  measuring and recording  the
volume or weight.
  4.3  Analysis. Record the data required on
the example sheet' shown  in Figure  17-4.
Handle each sample container as follows:
  Container No. 1. Leave the contents in the
shipping container or transfer the filter and
any loose particulate from the sample con-
tainer to a tared glass weighing dish. Desic-
cate for 24 hours in a desiccator  containing
anhydrous calcium sulfate. Weigh to a con-
stant weight and  report the results to the
nearest 0.1 mg. For purposes of this Section,
4.3, the term "constant weight" means a dif-
ference of no more than 0.5  mg or 1 percent
of total weight less tare weight, whichever is
greater, between two consecutive  weighings,
with no less  than 6 hours of desiccation
time between weighings.
  Alternatively, the sample may be  oven
dried at the average stack  temperature or
105° C (220° F), whichever is less, for 2 to 3
hours, cooled in the desiccator, and weighed
to a constant weight, unless  otherwise speci-
fied by the Administrator.  The tester may
also opt to oven dry the sample at the aver-
age stack temperature  or 105° C (220°  F),
whichever is less, for 2 to 3 hours, weigh the
sample, and  use this  weight as a  final
weight.
                                        B-71

-------
 Chapter I—Environmental Protection Agency
 H»nt__	
 Date	
                            App. A.
 Run No..
 Filter No.
Amount liquid lost during transport
Acttone blank volume, ml	
Acetone wash volume, ml	
Aettone Mack concentration, mg/mg (equation 17-4)
Acetone wash blank, mg (equation 17-5J  	
CONTAINER
NUMBER
1
2
TOTAL
WEIGHT OF PARTICULATE COLLECTED.
mg
FINAL WEIGHT


Z^^^^d
TARE WEIGHT


^xd
Less acetone blank
Weight of particulate matter
WEIGHT GAIN






FINAL
INITIAL
LIQUID COLLECTED
TOTAL VOLUME COLLECTED
VOLUME OF LIQUID
WATER COLLECTED
IMPINGER
VOLUME.
ml




SILICA GEL
WEIGHT,
g



fT" ml
    * CONVERT WEIGHT OF WATER TO VOLUME BY DIVIDING TOTAL WEIGHT
      INCREASE BY DENSITY OF WATER (1g/ml).
Figure 17-4. Analytical data.
INCREASE.,  8VOLUMEWATB|
   1 g/ml
                               B-72

-------
App. A

  Container No. 2. Note the level of liquid in
the container and confirm on the analysis
sheet  whether  or not  leakage occurred
during transport. If a noticeable amount of
leakage has occurred, either void the sample
or use methods, subject to the approval of
the Administrator, to  correct the final re-
sults. Measure the liquid in  this container
either volumetrically to ±1 ml or gravime-
trically to ±0.5 g. Transfer the contents to a
tared 250-ml beaker and evaporate to dry-
ness at ambient temperature and pressure.
Desiccate for 24 hours and weigh to a con-
stant weight. Report the results to the near-
est 0.1 mg.
  Container No. 3.  This step may be con-
ducted in the field. Weigh the spent silica
gel (or silica gel plus impinger) to the near-
est 0.5 g using a balance.
  "Acetone Blank" Container. Measure ac-
etone in this container either volumetrically
or gravimetrically. Transfer the acetone to a
tared 250-ml beaker and evaporate to dry-
ness at ambient temperature and pressure.
Desiccate for 24 hours and weigh to a con-
stant weight. Report the results to the near-
est 0.1 mg.
  NOTE: At the option of the tester, the con-
tents of Container No. 2 as well  as the ac-
etone blank container may be evaporated at
temperatures higher than ambient. If evap-
oration is done at an elevated temperature,
the temperature must be below the boiling
point of the solvent; also, to prevent "bump-
ing," the evaporation process must be close-
ly supervised, and  the  contents  of  the
beaker must be swirled occasionally to main-
tain an even temperature. Use extreme care,
as acetone is highly  flammable  and  has a
low flash point.
  5.  Calibration. Maintain a laboratory log
of all calibrations.
  5.1  Probe  Nozzle. Probe nozzles shall be
calibrated before  their  initial use  in the
field.  Using  a  micrometer, measure  the
inside diameter of the nozzle to the nearest
0.025 mm (0.001 in.).  Make  three separate
measurements  using  different  diameters
each time, and obtain the average of the
measurements. The difference between the
high and low numbers shall not  exceed 0.1
mm  (0.004  in.).  When  nozzles  become
nicked, dented,  or corroded, they shall be
reshaped,   sharpened,  and  recalibrated
before use. Each nozzle shall be permanent-
ly and uniquely identified.
  5.2  Pitot Tube. If the pitot tube is placed
in an interference-free arrangement with re-
spect  to the other probe assembly compo-
nents, its baseline (isolated tube)  coefficient
shall be determined as outlined in Section 4
of Method 2. If the probe assembly is not in-
terference-free,  the pitot tube assembly co-
efficient shall be determined by calibration,
using methods  subject to the approval  of
the Administrator.
   Title 40—Protection of Environment

  5.3  Metering System. Before its initial
use in the field, the metering system shall
be calibrated according to the procedure
outlined in APTD-0576. Instead of physical-
ly adjusting the dry gas meter, dial readings
to correspond to the wet test meter read-
ings,  calibration  factors may be used to
mathematically correct the gas  meter dial
readings to the proper values.
  Before calibrating the metering system, it
is suggested that a leak-check be conducted.
For  metering systems having  diaphragm
pumps, the  normal leak-check procedure
will not detect leakages within  the pump.
For  these cases  the  following  leak-check
procedure is suggested: make a 10-minute
calibration  run at  0.00057  m'/min (0.02
cfm); at the end of the run, take the differ-
ence  of the  measured wet test  meter and
dry gas meter volumes; divide the difference
by 10, to  get the leak rate. The leak rate
should not   exceed  0.00057  m'/min (0.02
cfm).
  After each field use, the calibration of the
metering  system shall be checked by per-
forming three calibration runs at a single,
intermediate orifice setting (based  on  the
previous field test), with the vacuum set at
the maximum value reached during the test
series. To adjust the vacuum,  insert a valve  .
between the wet test meter and the inlet of
the metering system. Calculate the average
value of the calibration factor. If the cali-
bration has  changed by more than 5 per-
cent, recalibrate the  meter  over the  full
range of orifice  settings,  as outlined in
APTD-0576.
  Alternative procedures, e.g., using the ori-
fice meter coefficients, may be used, subject
to the approval of'the Administrator.

  NOTE: If  the dry gas  meter coefficient
values  obtained  before and  after a  test
series differ by more  than 5 percent,  the
test series shall either be voided, or calcula-
tions for  the test series shall be performed
using whichever  meter  coefficient value
(i.e., before or after) gives the  lower value of
total sample volume.
  5.4  Temperature Gauges. Use the proce-
dure in Section 4.3 of Method 2 to calibrate
in-stack temperature gauges. Dial thermom-
eters, such as are used for the dry gas meter
and condenser  outlet, shall  be  calibrated
against mercary-in-glass thermometers.
  5.5  Leak  Check  of Metering   System
Shown in Figure 17-1. That portion of the
sampling  train from the pump to the orifice
meter should be leak checked  prior to initial
use and after each shipment.  Leakage after
the pump will result in less volume being re-
corded than is actually sampled. The follow-
ing procedure is suggested (see Figure 17-5).
Close the main valve on the  meter box.
Insert  a  one-hole  rubber  stopper  with
rubber tubing attached into the orifice ex-
                                         B-73

-------
Chapter I—Environmental Protection Agency
haust pipe. Disconnect and vent the low side
of the orifice manometer. Close off the low
side orifice tap. Pressurize the system to  13
to 18 cm (5 to 7 in.) water column by blow-
ing into the rubber tubing. Pinch off the
                               App. A
tubing and observe the manometer for one
minute.  A loss  of  pressure on the mano-
meter indicates a leak in the meter box;
leaks, if present, must be corrected.
                                    B-74

-------
             RUBBER
             TUBING
                        RUBBER      ADICIPE
                       STOPPER      °RIFICE
                                                                       VACUUM
                                                                       GAUGE
CO

en
 BLOW INTO TUBING
 UNTIL MANOMETER
READS 5 TO 7 INCHES
  WATER COLUMN
                     ORIFICE
                   MANOMETER
   MAIN VALVE
    CLOSED

AIR-TIGHT
  PUMP
                                   Figure 17-5. Leak check of meter box.
                                                                                                   a
                                                                                                   5'
                                                                                                   f
                                                                                                   i

-------
 Chapter I—Environmental Protection Agency
                                App.A
  5.6  Barometer. Calibrate  against a mer-
 cury barometer.
  6. Calculations. Carry out calculations, re-
 taining at least  one  extra  decimal  figure
 beyond that of the acquired data. Round off
 figures  after the final calculation.  Other
 forms of the equations may be used as long
 as they give equivalent results.
  6.1  Nomenclature.
 A,=Cross-sectional area of nozzle, m2 (ft2).
 B.,=Water vapor in the gas stream, propor-
   tion by volume.
 C.=Acetone blank residue  concentration,
   mg/g.
 c,=Concentration of participate matter in
   stack gas, dry basis, corrected to  stand-
   ard conditions, g/dscm (g/dscf).
 I=Percent of isokinetic sampling.
 L,=Maximum  acceptable  leakage  rate for
   either a pretest leak check or for a leak
   check following a component change;
   equal to 0.00057 m'/min (0.02 cfm) or 4
   percent  of the  average  sampling rate,
   whichever is less.
 L,=Indivldual leakage rate observed during
   the leak check conducted prior to the
   "!"•" component change (1=1, 2, 3 ... n),
   mVmin(cfm).,
L,=Leakage rate  observed during the post-
   test leak check, m Vmin (cfm).
m,,=Total amount of particulate matter col-
   lected, tng.
Mw=Molecular weight of water, 18.0 g/g-
   mole (18.0 Ib/lb-mole).
m.=Mass of residue of acetone after evapo-
   ration, mg.
Ptar=Barometric pressure  at the sampling
   site, mm Hg (in. Hg).
P,=Absolute stack gas pressure, mm Hg (in.
   Hg).
P.u«=Standard  absolute pressure, 760 mm
   Hg (29.92 in. Hg).
R=Ideal gas constant, 0.06236 mm Hg-mV
   •K-g-mole (21.85 in. Hg-ft V°R-lb-mole).
T.,—Absolute average dry gas  meter tem-
   perature (see Figure 17-3), °K (°R).
T.=Absolute average stack gas temperature
   (see Figure 17-3), °K OR).
T,u=Standard  absolute temperature,  293°K
   (528'R).
V«=Volume of acetone blank, ml.
V.w=Volume of acetone used in wash, ml.
Vic—Total volume of liquid collected  in im-
   pingers and silica gel (see Figure 17-4),
   ml.  .
V«,=Volume of gas sample as measured by
   dry gas meter, dcm (dcf).
Vm(,u)=Volume of gas sample measured by
   the dry gas meter,  corrected to standard
   conditions, dscm (dscf).
V.(,ui)=Volume of water vapor }n  the gas
   sample,  corrected  to  standard  condi-
   tions, scm (scf).
v,«Stack gas velocity, calculated by Method
   2, Equation  2-9,  using  data obtained
   from Method 17, m/sec (ft/sec).
W.=Weight of residue in acetone wash, mg.
Y=Dry gas meter calibration coefficient.
AH=Average pressure  differential  across
   the orifice meter (see Figure 17-3), mm
   H»O (in. HaO).
p,=Density of acetone, mg/ml (see label on
   bottle).
»w=Density of water, 0.9982 g/ml (0.002201
   Ib/ml).
8- Total sampling time, min.
0,=Sampling time interval, from the begin-
   ning of a run until the first component
   change, min.
#,=Sampling time  interval,  between  two
   successive component  changes,  begin-
   ning with the interval between the first
   and second changes, min.
0B=Sampling time interval, from the final
   (n"1) component change until the end of
   the sampling run, min.
13.6=Specific gravity of mercury.
60=Sec/min.
100=Conversion to percent.
  6.2  Average dry  gas  meter temperature
and average orifice  pressure drop. See data
sheet (Figure 17-3).
  6.3  Dry Gas Volume.  Correct the sample
volume measured by the dry gas meter to
standard conditions (20° C, 760 mm Hg or
68° F, 29.92 in. Hg)  by using Equation 17-1.
   Vm(std) = Vin
                          Equation 17-1
where:

K,=0.3858°  K/mm Hg for  metric  units;
   17.64' R/in. Hg for English units.

  NOTE: Equation 17-1 can be used as writ-
ten unless the leakage rate  observed during
any of the mandatory leak  checks (i.e., the
post-test leak check or leak  checks conduct-
ed prior to component changes) exceeds L,.
If LP or L, exceeds L., Equation 17-1 must be
modified as follows:
  (a) Case I. No component changes made
during sampling run.  In this case, replace
Vm in Equation 17-1 with the expression:
                                        B-76

-------
App. A
                                               Title 40—Protection of Environment
                                   -
  (b)  Case  II.  One  or more  component
changes made during the sampling run. In
this case, replace Vm in Equation 17-1 by the
expression:
                            L1 ' La>
                        ' 
-------
 Chapter I—Environmental Protection Agency

 K.=4.320 for metric units; 0.09450 for Eng-
    lish units.

  6.12  Acceptable  Results.  If  90  percent
 
-------
         APPENDIX C



EPA REFERENCE METHODS 16..16A
              C-l

-------

-------
 APR. A
   Title 40—Protection of Environment
METHOD  16—SEMICONTINUOUS  DETERMINA-
  TION OF SULFUR EMISSIONS PROM STATION-
  ARY SOURCES

              Introduction

  The method described below  uses the
principle of gas chromatographic separation
and  flame  photometric  detection.  Since
there are many systems or sets of operating
conditions that represent usable methods of
determining sulfur emissions, all systems
which employ this principle, but differ only
In details of equipment and operation, may
be used  as  alternative methods, provided
that the criteria set below are met.
  1. Principle and Applicability.
  1.1  Principle. A gas sample is  extracted
from the emission source and diluted with
clean dry air. An aliquot of the diluted
 sample is then analyzed for hydrogen sul-
 fide (HjS),  methyl mercaptan (MeSH), di-
 methyl sulfide (DMS) and  dimethyl disul-
 fide (DMDS) by gas chromatographic 
-------
Chapter I—Environmental Protection Agency
                                APR. A
from the sample. In the example system,
SO, is removed by a citrate buffer solution
prior to GC injection. This scrubber will be
used when  SO, levels are high enough to
prevent baseline  separation from the re-
duced sulfur compounds.
  Compliance with this section can be dem-
onstrated by submitting chromatographs of
calibration  gases  with  SO, present  in  the
same quantities expected from the emission
source  to  be tested.  Acceptable systems
shall show baseline separation with the  am-
plifier attenuation set so that  the reduced
sulfur compound  of concern is at least 50
percent of full scale. Base line separation is
defined as a return to zero ± percent in the
interval between peaks.
  4. Precision and Accuracy.
  4.1 OC/PPD and Dilution System Cali-
bration Precision.  A series of three consecu-
tive injections of  the same calibration  gas,
at any  dilution, shall produce results which
do not  vary by more than  ±5 percent from
the mean of the three injections.
  4.2 GC/FPD and Dilution System Cali-
bration Drift. The calibration drift deter-
mined  from the mean of three injections
made at the  beginning and end of any 8-
hour period shall not exceed ± percent.
  4.3 System  Calibration Accuracy. Losses
through the sample transport system must
be measured and  a correction factor devel-
oped to adjust the calibration accuracy to
100 percent.
  5. Apparatus (See Figure 16-1).
  5.1. Sampling.
  6.1.1  Probe. The probe must be made of
inert material such  as stainless steel or
glass. It should be designed to incorporate a
filter and to allow calibration  gas to enter
the probe at or near the sample entry point.
Any portion of the probe not exposed to the
stack gas must be heated  to prevent mois-
ture condensation.
  5.1.2  Sample Line. The sample line must
be made of Teflon,' no greater than 1.3 cm
(V4)  Inside  diameter.  All  parts from  the
probe to the  dilution system must be ther-
mostatically heated to 120° C.
  5.1.3  Sample Pump. The sample pump
shall be a leakless Teflon-coated diaphragm
type or equivalent. If the pump is upstream
of the dilution system, the pump head must
be heated to 120* C.
  5.2 Dilution System. The dilution system
must be constructed such that all  sample
contacts are made of inert materials (e.g.,
stainless steel or Teflon). It must be heated
to 120* C. and be capable of approximately a
9:1 dilution of the sample.
  5.3 SO, Scrubber. The SO, Scrubber is a
midget impinger packed with glass wool to
  'Mention of trade names or specific prod-
ucts does not constitute endorsement by the
Environmental Protection Agency.
eliminate entrained mist and charged with
potassium citrate-citric acid buffer.
  5.4  Gas Chromatograph. The gas chro-
matograph must have at least the following
components:
  5.4.1 Oven.  Capable of  maintaining the
separation column at the proper operating
temperature ±1° C.
  5.4.2 Temperature  Gauge.  To  monitor
column oven,  detector,  and exhaust tem-
perature ±1° C.
  5.4.3 Flow System. Gas metering system
to measure  sample, fuel,  combustion  gas,
and carrier gas flows.
  5.4.4 Flame Photometric Detector.
  5.4.4.1  Electrometer. Capable of full scale
amplification of linear ranges of 10"9to 10~4
amperes full scale.
  5.4.4.2  Power Supply. Capable of deliver-
ing up to 750 volts.
  5.4.4.3  Recorder.   Compatible  with  the
output voltage range of the ele'ctrometer.
  5.5  Gas  Chromatograph  Columns.  The
column system must be  demonstrated to be
capble of resolving  the four major reduced
sulfur compounds:  HaS, MeSH, DMS,  and
DMDS. It must also demonstrate  freedom
from known interferences.
  To demonstrate that adequate resolution
has been achieved, the tester must submit a
Chromatograph of a calibration gas contain-
ing all four of the  TRS compounds in the
concentration range of the applicable stand-
ard. Adequate resolution will be defined as
base line separation of adjacent peaks when
the amplifier attenuation is set so  that the
smaller peak is at  least 50 percent of full
scale. Base line separation is defined in Sec-
tion 3.4. Systems not meeting this criteria
may be considered  alternate methods  sub-
ject to the approval of the Administrator.
  5.6  Calibration System. The calibration
system must contain the  following compo-
nents.
  5.6.1 Tube Chamber. Chamber of glass or
Teflon of sufficient dimensions  to house
permeation tubes.
  5.6.2 Flow System. To measure air  flow
over permeation tubes at ±2 percent. Each
flowmeter shall be calibrated after a com-
plete  test series with a wet test meter. If the
flow measuring device differs from the wet
test meter by  5 percent, the completed test
shall  be discarded. Alternatively, the tester
may elect to use the flow data that would
yield  the lower flow measurement. Calibra-
tion with a wet test meter before a test is
optional.
  5.6.3  Constant Temperature Bath. Device
capable  of maintaining  the  permeation
tubes at the calibration temperature within
±0.1' C.
  5.6.4 -Temperature Gauge. Thermometer
or equivalent  to monitor bath temperature
within ±r C.
  6. Reagents.
                                        C-4

-------
 APR. A
   Title 40—Protection of Environment
  6.1 Fuel.  Hydrogen   (Hi)   prepurified
 grade or better.
  6.2 Combustion Gas. Oxygen (O.) or air,
 research purity or better.
  6.3 Carrier Gas.  Prepurified grade  or
 better.
  6.4 Diluent. Air containing less than 50
 ppb total sulfur compounds and less than 10
 ppm each  of moisture and  total hydrocar-
 bons. This gas  must be heated prior to
 mixing with the sample to avoid water con-
 densation at the point of contact.
 • 6.5 Calibration Gases. Permeation tubes,
 one each of HiS, MeSH, DMS, and DMDS,
 agravimetrically calibrated and  certified at
 some convenient operating  temperature.
 These tubes consist of hermetically sealed
 PEP Teflon tubing in which a liquified gas-
 eous substance is enclosed. The enclosed gas
 permeates through the tubing wall at a con-
 stant rate. When the temperature is con-
 stant, calibration gases  Governing a wide
 range of known  concentrations can be gen-
 erated by varying and accurately measuring
 the flow rate of diluent gas passing over the
 tubes. These calibration gases  are used to
 calibrate the GC/PPD system arid the dilu-
 tion system.
  6.6 Citrate Buffer. Dissolve 300 grams of
 potassium  citrate and 41 grams of anhy-
 drous citric acid in 1 liter of deionized water.
 284 grams of sodium citrate may be substi-
 tuted for the potassium citrate.
  7. Pretest Procedures. The following proce-
 dures are optional but would be helpful in
 preventing any problem which might occur
 later and invalidate the entire test.
  7.1 After  the  complete  measurement
system  has been set up at the site arid
 deemed to be operational, the following pro-
cedures should be  completed before sam-
pling is initiated.
  7.1.1 Leak Test.   Appropriate leak test
procedures should be employed to verify the
integrity of all  components, sample  lines,
and connections. The following leak test
procedure is suggested: For components up-
stream  of  the  sample  pump,  attach  the
probe end of the sample line to a ma-  no-
meter or vacuum gauge, start the pump and
pull greater than 50  mm (2 in.) Hg vacuum,
close off the pump outlet, and then stop the
pump and ascertain that there is no leak for
 1 minute. For components after the pump,
apply a slight positive pressure and check
for leaks by applying a liquid (detergent in
water, for example) at each joint. Bubbling-
indicates the presence of a leak.
  7.1.2  System  Performance.   Since  the
complete system is calibrated following each
test, the precise calibration of each compo-
nent is not critical. However, these compo-
nents should be verified to be operating
properly. This verification can be performed
by observing the response of flowmeters or
 of the GC output to changes in flow rates or
 calibration  gas concentrations and  ascer-
 taining the response to be within predicted
 limits. In any component, or if the complete
 system fails to respond in a normal and pre-
 dictable  manner, the source of the discrep-
 ancy should be identified and corrected
 before proceeding.
  8. Calibration. Prior to any sampling run,
 calibrate the system using the following
 procedures. (If more than one run is per-
 formed during any 24-hour period, a calibra-
 tion need not  be performed  prior  to  the
 second and any subsequent runs. The cali-
 bration must, however, be  verified as pre-
 scribed in Section 10,  after the last  run
 made within the 24-hour period.)
  8.1 General Considerations.  This section
 outlines  steps to be followed for use of the
 GC/FPD and the dilution system. The pro-
 cedure does not include detailed  instruc-
 tions because the operation of these systems
 is complex,  and it requires a understanding
 of the individual system  being used. Each
 system should  include a  written operating
 manual  describing in detail the operating
 procedures associated with each component
 in the measurement systerii. In addition, the
 operator should be familiar with the operat-
 ing principles of the components; particular-
 ly the GC/FPD. The citations in the Bib-
 liography at the end of this method are  rec-
 ommended for review for this purpose.
  8.2 Calibration Procedure. Insert the per-
 meation   tubes  into the  tube chamber.
 Check the bath  temperature  to  assure
 agreement with the calibration  temperature
 of the tubes within ±0.1°  C. Allow 24 hours,
 for the  tubes to equilibrate. Alternatively
 equilibration may be verified by injecting
 samples  of calibration gas at 1-hour inter-
 vals. The permeation tubes can be assumed
 to have reached, equilibrium when consecu-
 tive hourly samples agree within the preci-
 sion limits of Section 4.1.
  Vary the amount of air flowing over  the
 tubes to  produce the desired concentrations
 for  calibrating  the analytical  and dilution
systems.  The air flow across the tubes must
 at all times exceed the flow requirement of
 the analytical systems. The concentration in
parts per million generated by a tube con-
 taining a specific permeant can be calculat-
ed as follows:
              C  =  K
where:
                          Equation 16-1
                                        C-5

-------
APP.A
   Till* 40—Protection of Environment
  10.2  Recallbratlon. After each run, or
after a series of runs made within a 24-hour
period, perform a partial recalibration using
the procedures in Section 8. Only HjS (or
other permeant) need be used to recalibrate
the GC/FPD analysis system (8.3) and the
dilution system (8.5).
  10.3  Determination of Calibration Drift.
Compare  the calibration curves obtained
prior to the runs, to the calibration  curves
obtained under paragraph 10.1. The calibra-
tion drift should not exceed the  limits set
forth in subsection 4.2. If the drift exceeds
this limit, the  intervening run  or runs
should be considered not valid. The  tester,
however, may instead have the  option of
choosing  the calibration data set  which
would give the highest sample values.
  11. Calculations.
  11.1  Determine the  concentrations  of
each reduced sulfur compound detected di-
rectly from the calibration  curves. Alterna-
tively, the concentrations may be calculated
using the'equation for the least square line.
  11.2 Calculation of TRS. Total reduced
sulfur will be determined for each anaylsis
made by  summing the  concentrations of
each  reduced sulfur  compound  resolved
during a given analysis.

   TRS=2 (H.S. MeSH, DMS, 2DMDS)d
                          Equation 16-2
where:
TRS=Total  reduced  sulfur in  ppm,  wet
   basis.
HiS-Hydrogen sulfide, ppm.
MeSH=Methyl mercaptan. ppm.
DMS - Dimethyl sulfide, ppm.
DMDS=Dimethyl disulfide, ppm.
d-Dilution factor, dimensionless.
  11.3  Average TRS. The average TRS will
be determined as follows:
Average TRS=
       Average TRS=
Average TRS=Average total reduced suflur
   in ppm, dry basis.
TRS|=Total reduced sulfur in ppm as deter-
   mined by Equation 16-2.
N=Number of samples.
B».=Praction  of volume of water vapor in
   the gas stream as determined by refer-
   ence method 4—Determination of Mois-
   ture in Stack Gases (36 PR 24887).
  11.4 Average concentration of individual
reduced sulfur compounds.
                          Equation 16-3
where:

S,=Concentration  of any  reduced  sulfur
   compound from  the  ith  sample injec-
   tion, ppm.
C=Average concentration of any one of the
   reduced sulfur compounds for the entire
   run, ppm.
N=Number of injections  in any-run period.
  12. Example System. Described below is a
system utilized by  EPA in gathering NSPS
data. This system  does not now reflect all
the latest developments  in equipment and
column technology,  but  it does represent
one system that has been demonstrated to
work.
  12.1  Apparatus.
  12.1.1 Sampling  System.
  12.1.1.1  Probe. Figure 16-1 illustrates the
probe used in lime kilns  and  other sources
where  significant  amounts  of  particulate
matter are present,  the  probe is designed
with the deflector shield placed between the
sample and the gas inlet holes and the glass
wool plugs to reduce clogging of the filter
and possible adsorption of sample gas. The
exposed portion of the probe between the
sampling port and  the sample line is heated
with heating tape.
  12.1.1.2  Sample Line 9U inch inside diam-
eter Teflon tubing, heated  to 120° C. This
temperature is controlled by a thermostatic
heater.
  12.1.1.3  Sample  Pump. Leakless Teflon
coated diaphragm  type or equivalent. The
pump head is heated to 120° C by enclosing
it  in  the  sample dilution  box  (12.1.2.4
below).
  12.1.2  Dilution System. A schematic dia-
gram  of the  dynamic dilution  system  is
given in Figure 16-2. The dilution system is
constructed such that all sample  contacts
are made of  inert materials. The  dilution
system which is heated to 120° C must be ca-
pable  of a  minimum of 9:1 dilution of
sample.  Equipment  used in   the  dilution
system is listed below:
  12.1.2.1  Dilution Pump.  Model  A-150
Kohmyhr  Teflon  positive  displacement
type, nonadjustable  150  cc/min. ±2.0 per-
cent, or equivalent, per dilution stage. A 9:1
dilution of sample  is accomplished  by com-
bining 150  cc of sample with 1,350 cc of
clean dry air as shown in Figure 16-2.
  12.1.2.2  Valves.  Three-way  Teflon sole-
noid or manual type.
                                       C-6

-------
 Chapter I—Environmental Protection Agency
                                 App. A
  12.1.2.3 Tubing.  Teflon tubing and  fit-
 tings are used throughout from the sample
 probe to the GC/PPD to present an" inert
 surface for sample gas.
  12.1.2.4 Box.  Insulated box, heated and
 maintained  at 120" C, of sufficient dimen-
 sions to house dilution apparatus.
  12.1.2.5 Plowmeters.   Rotameters    or
 equivalent to measure flow  from 0 to 1500
 ml/mm ±1 percent per dilution stage.
  12.1.3 SO2 Scrubber. Midget impinger with
 15 ml of potassium citrate buffer to absorb
 SO, in the sample.
  12.1.4  Gas   Chromatograph  Columns.
 Two types of columns are used for separa-
 tion  of  low and  high molecular weight
 sulfur compounds:
  12.1.4.1 Low  Molecular Weight Sulfur
 Compounds Column (GC/FPD-I).
  12.1.4.1.1  Separation Column. 11 m by
 2.16 mm (36 ft by 0.085 in) inside diameter
 Teflon tubing  packed with  30/60 .mesh
 Teflon coated with 5 percent  polyphenyl
 ether  and  0.05  percent orthophosphoric
 acid, or equivalent (see Figure 16-3).
  12.1.4.1.2  Stripper or Precolumn.  0.6  m
 by 2.16 mm (2 ft by 0.085 in) inside diameter
 Teflon tubing.
  12.1.4.1.3  Sample Valve.  Teflon  10-port
 gas sampling valve, equipped with a 10 ml
 sample loop, actuated by compressed  air
 (Figure 16-3).
  12.1.4.1.4  Oven.  For containing  sample
 valve,  stripper   column  and  separation
 column.  The  oven should  be  capable  of
 maintaining an elevated temperature rang-
 ing from ambient to 100° C, constant within
 ±r c.
  12.1.4.1.5  Temperature Monitor. Thermo-
 couple pyrometer to measure column oven,
 detector, and exhaust temperature ±1' C.
  12.1.4.1.6  Flow  System. . Gas metering
 system to measure sample flow, hydrogen
 flow, and oxygen flow (and nitrogen carrier
 gas flow).
  12.1.4.1.7  Detector.  Flame  photometric
 detector.
  12.1.4.1.8  Electrometer. Capable  of  full
 scale  amplification  of linear ranges of 10"s
 to 10"' amperes full scale.
  12.1.4.1.9  Power Supply. Capable of deli-
 vering up to 750 volts.
  12.1.4:1.10 Recorder.  Compatible  with
 the output voltage range of the electrom-
 eter.
  12.1.4.2  High  Molecular  Weight  Com-
 pounds Column (GC/FPD-II).
  12.1.4.2.il  Separation Column. 3.05 m by
 2.16 mm (10 ft by 0.0885 in) inside diameter
 Teflon  tubing  packed .with  30/60  mesh
 Teflon coated with 10 percent Triton X-305,
 or equivalent.
  12.1.4.2.2  Sample Valve. Teflon 6-port gas
sampling  valve  equipped  with  a  10  ml
sample loop, actuated  by compressed  air
 (Figure 16-3).
  12.1.4.2.3  Other Components. All compo-
nents same, as in 12.1.4.1.5 to 12.1.4.1.10.
  12.1.5  'Calibration.    Permeation   tube
system (figure 16-4).
  12.1.5.1 Tube Chamber. Glass  chamber
of sufficient dimensions to  house perme-
ation tubes.
  12.1.5.2 Mass  Flowmeters.  Two  mass
flowmeters in the range 0-3 1/min. and 0-10
1/min. to measure air flow over permeation
tubes at  ±2 percent. These flowmeters shall
be cross-calibrated at the beginning of each
test. Using  a convenient flow rate in the
measuring range of both  flowmeters,  set
and monitor the flow rate of gas over the
permeation  tubes. Injection  of calibration
gas generated at this flow rate as measured
by one flowmeter followed by injection  of
calibration  gas at  the  same  flow rate  as
measured by the other flowmeter should
agree within the specified precision limits.
If they do not, then 'there is a problem with
the mass flow  measurement. Each  mass
flowmeter shall be calibrated prior to the
first test with a wet test meter and thereaf-
ter, at least once each year.
  12.1.5.3 Constant Temperature Bath. Ca-
pable of maintaining permeation tubes  at
certification temperature of  30°  C. within
±0.1° C.
  12.2  Reagents
  12.2.1   Fuel.  Hydrogen (H,) prepurified
grade or better.
  12.2.2.  Combustion Gas. Oxygen (O3) re-
search purity or better.
  12.2.3   Carrier Gas. Nitrogen (N2) prepuri-
fied grade or better.
  12.2.4   Diluent. Air  containing less than
50 ppb total sulfur compounds and less than
10 ppm each of moisture and total hydro-
carbons,  and  filtered   using  MSA  filters
46727 and 79030, or equivalent. Removal  of
sulfur compounds can be verified by inject-
ing dilution air  only, described in Section
8.3.
  12.2.5   Compressed Air. 60  psig for GC
valve actuation.
  12.2.6   Calibrated   Gases.  ' Permeation
tubes gravimetrically  calibrated and certi-
fied at 30.0° C.
  12.2.7   Citrate Buffer. Dissolve 300 grams
of potassium -citrate and 41 grams of anhy-
drous citric acid in 1 liter of deionized water.
284 grams of sodium citrate may be substi-
tuted for the potassium citrate.
  12.3  Operating Parameters.
  12.3.1   Low-Molecular  Weight  Sulfur
Compounds. The operating parameters for
the GC/FPD system used for low molecular
weight compounds are as follows: nitrogen
carrier gas flow rate of 50 cc/min, exhaust
temperature of 110° C, detector temperature
of 105° C, oven temperature of 40° C, hydro-
gen flow rate of 80 cc/min, oxygen flow rate
of 20 cc/min, and sample flow rate between
20 and 80 cc/min.
                                       C-7

-------
App. A

  12.3.2 High-Molecular  Weight „ Sulfur
Compounds. The operating parameters for
the GC/FPD  system for high  molecular
weight compounds are the same as in 12.3.1
except: oven temperature of 70° C, and ni-
trogen carrier gas flow of 100 cc/min.
  12.4  Analysis Procedure.
  12.4.1 Analysis.   Aliquots   of  diluted
sample are injected  simultaneously  into
both GC/PPD analyzers  for analysis. GC/
FPD-I is used to measure the low-molecular
weight reduced sulfur compounds. The low
molecular weight compounds Include hydro-
gen  sulfide,  methyl  mercaptan,  and di-
methyl sulfide. GC/FPD-II is used to re-
solve the high-molecular weight compound.
The high-molecular weight'compound is di-
methyl dlsulf ide.
  12.4.1.1  Analysis    of   Low-Molecular
Weight Sulfur Compounds.  The  sample
valve is actuated for 3 minutes in which
time an aliquot of diluted sample is injected
into the  stripper column and  analytical
column. The valve is then deactivated for
approximately  12 minutes In  which time.
the analytical column continues to be fore-
flushed, the stripper column Is backflushed,
and the sample loop is refilled. Monitor the
responses. The elution time for  each com-
pound will be determined during calibra-
tion.
  12.4.1.2  Analysis    of    High-Molecular
Weight Sulfur Compounds. The procedure
IB essentially the same as above except that
no stripper column Is needed.
  13. Bibliography.
   Title 40—Protection of Environment

   13.1  O'Keeffe, A. E. and G. C. Ortman.
 "Primary Standards for Trace Gas Analy-
 sis."  Analytical Chemical  Journal,  38,760
 (1966).
   13.2  Stevens, R. K., A. E. O'Keeffe, and
. G. C. Ortman.  "Absolute Calibration of a
 Flame Photometric  Detector  to Volatile
 Sulfur Compounds at Sub-Part-Per-Million
 Levels." Environmental Science and Tech-
 nology, 3:7 (July, 1969).
   13.3  Mulick, J.  D., B. K. Stevens, and R.
 Baumgardner.  "An Analytical  System De-
 signed to  Measure  Multiple  Malodorous
 Compounds Related  to Kraft Mill  Activi-
 ties." Presented at the 12th Conference on
 Methods in Air Pollution and Industrial Hy-
 giene Studies, University of Southern Cali-
 fornia, Los Angeles, CA. April 6-8,1971.
   13.4  Devonald,  R. H., R. S. Serenius, and
 A. D. Mclntyre. "Evaluation of the Flame
 Photometric Detector for Analysis of Sulfur
 Compounds." Pulp and Paper Magazine  of
 Canada, 73,3 (March, 1972).
   13.5  Grimley, K. W., W. S. Smith, and R.
 M. Martin. "The Use of a Dynamic Dilution
 System in the Conditioning of Stack Gases
 for Automated Analysis by a Mobile Sam-
 pling Van." Presented at the 63rd Annual
 APCA Meeting in St. Louis, Mo. June 14-19,
 1970.
   13.6 General Reference. Standard Meth-
 ods of Chemical Analysis Volume III A and
 B Instrumental  Methods. Sixth  Edition.
 Van Nostrand Reinhold Co.
                                         C-8

-------
Chapter I—Environmental Protection Agency
App. A
                                                                      •o
                                                                      10


                                                                      5
                                                                      f.
                                                                      to
                                                                      i
                                                                      o
                                                                      u

                                                                      M

                                                                      cn

                                                                      «>



                                                                      I-
                                                                      in
                                                                      a>



                                                                     1
                                                                     0.
                                    C-9

-------
                                                                                        TOCC/FPDAIALYZEM

                                                                                          19:1
                              FILTER
                            (CLASS WOOL)
O
(-»
O
                                                                                        •p
                                                                                         >
                                                                                                                                            FLOVVMETER
DIAPHRAGM
  PUMP
 (HEATED)
                            Figure 16-2.Sampl!ng and dilution appiratui
    DILUTION BOX HEATED
          TOIOfl'C

VEMT
                                                                                                                                                         s.
                                                                                                                                                         m

                                                                                                                                                         I

-------
Chapter I—Environmental Protection Agency
App.A
                               C-ll

-------

                                                     TO raSTRUMEHTS
                                                          AND
                                                     DILUTION SYSTEM
O
I
                                            CONSTANT
                                           TEMPERATURE
                                              BATH
                                                                                                    	   DILUENT

                                                                                                    DRIER   ^-A0'g

                                                                                                              NITROGEN
                                                                                            GLASS
                                                                                           CHAMBER
                                                           PERMEATION
                                                              TUBE
 •p
 >
                                                            Figure 16-4. Apparatus for field calibration.
 i
 f
 I
                                                                                                                        8.
                                                                                                                        m
I

-------
Chapter I—Environmental Protection Agency
App.A
                                  OS
                                  0«*
                                              Itu

                                                             GC

                                                             X
               -j a.



               II
                                                                     I HI
                                                                     ;o
                                                                     • c
                                                                     '.9

                                                                      c

                                                                      e


                                                                      1
                                                                     o

                                                                     in
                                                                     a


                                                                     iE

SV3
NOI1VU8I1V3
                                      C-13

-------
                  Federal Register /  Vol. 46,  No. 117 /  Thursday, June 18, 1981 /  Proposed Rules
                                                                              31905
   1. By revising paragraph (d)(l) of
 f 60.285 to read as follows:

 5 60 285  Test methods and procedures.
                *     *
   (dj* " *
   (1) Method 18 or, at the discretion of
 Iho owner or operator, Method 16A for
 the concentration of TRS,
 *****
  2, By amending Appendix A by adding
 a new method as follows:   '
 Appendix A—Reference Methods
 *****

 Method 1GA. Determination of Total Reduced
 Sulfur Emissions From Stationary Sources
 (Impinger Technique)
  1. Applicability and Principle—1.1
Applicability. This is° an alternative method
 to Method 16 for determining total reduced
 sulfur (TRS) compounds from recovery
 furnaces, lime kilns,  and smelt dissolving
 tanks at kraft pulp mills. The TRS compounds
 include hydrogen sulflde, methyl mercaptan,
 dimethyl sulflde, dimethyl dlsulfide, and
 other reduced sulfur compounds (e.g.,
 cabonyl sulflde, if present). Therefore,
Method 16A might yield higher TRS
 concentrations than Method 16.
  The minimum detectable limit of the
method has been determined to be 0.04 ppm
TRS (compounds with single sulfur atom)
when sampling at 2 liters/min for 60 minutes.
 For an analytical accuracy of at least ±5
 percent, a minimum sulfur dioxide (SO?) mass
 of 500 ug should be collected. The upper
 concentration limit of the method generally
 exceeds all encountered TRS levels from
 kraft pulp mills.
  1.2  Principle. A gas sample is extracted
 from the sampling point in the stack. SO2 is
 selectively removed from the sample using a
 citrate buffer solution. Then reduced sulfur
 compounds are oxidized and analyzed as SOa
 using the barium-thorin titration procedure of
 Method 6.
   Z. Apparatus—2.1 Sampling. The sampling
 train is shown in Figure 16A-1. The apparatus
 is the same as listed in Method 6, except as
 listed below. Other designs 'are acceptable
 provided that the sampling system meets .the
 performance check of Section 5.
   2.1.1   SO2 Scrubber. Two midget impingers
 in series packed with glass wool to eliminate
 entrained mist and charged with potassium
 citrate-citric acid buffer.
   2.1.2   Combustion Tube. Quartz glass with
 an expanded combustion chamber of 22 to 25
 mm and at least 30.5 cm long. The tube ends
 shall have an outside diameter of about 6 mm
 to accept Teflon tubing or Swagelok fittings.
   2.1.3   Combustion Tube Furnace. A
• furnace of sufficient size to enclose the
 combustion chamber of the combustion tube
 with a temperature regulator capable of
 maintaining the temperature at 815 ±15°C.
   2.1.4   Rate Meters. Rotameters, or
 equivalent, capable of measuring flow rate to
 within 2 percent of the selected flow rate.
   2.1.5   Probe Brush. Nylon bristle brush
 with stainless steel wire handle. The brush
 shall be properly sized and shaped and of
 sufficient length to brush out the entire length
 of the probe.
   2,2  Sample Recovery. Same as in Method
 6, Section 2.2.
   2.3  Analysis. Same as in Method 6,
 Section 2.3, except a 10-ml buret with 0.1-ml
 graduations is required and the
 spectrophotorheter is not needed.
   3. Reagents—Unless otherwise indicated,
 all reagents must conform to the
 specifications established by the Committee
 on Analytical Reagents of the American
 Chemical Society. Where such specifications
                                                                                       are not available, use the best available
  3.1  Sampling. Th'e following reagents are
needed;
  3.1.1  Water. Same as Method 6, Section
3.1.1.
  3.1.2  Hydrogen Peroxide, 3 percent. Same
as Method 6, Section 3.1.3 (40 Ml is needed
per sample).
  3.1.3  Citrate Buffer. Dissolve 300 g of
potassium citrate (or 284 g of sodium citrate)
and 41 g of anhydrou citric acid in 1 liter of
deionized distilled water.
  3.1.4  Calibration Gas. Hydrogen sulflde in
nitrogen (30 to 50 ppm) stored in aluminum
cylinders. Verify the concentration by-
Method 11.
  3.1.5  Combustion Gas. Air or oxygen
containing less than 50 ppb total sulfur
compounds and less than 10 ppm total
hydrocarbons.
  3.2  Sample Recovery and Analysis.
Deionized distilled water (as in 3.1.1) and the
same reagents as in Method 6, Section 3,3,
are required.
  4. Procedure—4.1 Sampling.
  4.1.1  Preparation of Collection Train. For
the SOa scrubber, measure 20 ml citrate
buffer solution into each of'two midget
impingers with glass wool packed in topi For
the Method 6 part of the train, measure 20 ml
of 3 percent hydrogen peroxide into each of
the first two midget impingers. Leave the final
midget impinger dry. Assemble the train as
shown in figure 16A-1. Place the SO2
scrubber as close to the stack wall as
practical. Adjust the probe heater to a
temperature sufficient to prevent water
condensation. Maintain the oxidation furnace
at 815°C. Place crushed ice and water around
the impingers.
BILLING CODE 6560-26-M
                                                               C-14

-------
        «•*=[ HEATED
PROBE
r>

t—i
en
                              S02SCRUBBER
                                                                                         THERMOMETER
                                                             OXIDATION

                                                               TUBE
                                                                              IMPINGERS
                                                   TUBE FURNACE
I
jH1
                                                                            >


                                                                          "ICE BATH
                                                                             SILICA GEL

                                                                             DRYING TUBE
                                                                                    h
                                                              THERMOMETER


                                                                  ft
                                        Figure  16A-1. Sampling train
                                                                                                 NEEDLE

                                                                                                 VALVE
                                                                                          PUMP
                                                                                                             •a
                                                                                                             CD
                                                                                                             9.
                                                                                                             n
                                                                                         CD.

                                                                                         )_&

                                                                                         s°


                                                                                         co
                                                                                         CD
                                                                                                             TJ

                                                                                                             O

                                                                                                             CD

                                                                                                             CD

                                                                                                             P*
                                                                                                             CD

                                                                                                             CO

-------
                 Federal Register /  Vol. 46.  No.  117  / Thursday. June 18.  1981  / Proposed  Rules          31907
  4.1.2  Leak-Check Procedure. Same as
 Method 6. Section 4.1.2.
  4.1.3  Sample Collection. Same as Method
 6. Section 4.1.3, except for the following:
 Adjust the sample flow to a constant rate of
 approximately 2.0 lilers/min (±10 percent) as
 Indicated by the rotameter. Other constant
 flow rates may also be usgd provided its
 acceptability is checked as in Section 5.
 Collect the sample for 60 minutes. The 15-
 mlnulo purge of the train following collection
 need not be performed.
  In Method 16, a sample run is composed of
 10 individual analyses (injects) performed
 over a period of not less than 3 hours or more
 than 6 hours. For Method 16A to be
 consistent with Method 16, the following may
be used to obtain a sample run: (1) collect
 three GO-mfnute samples or (2) collect one 3-
hour sample with a total gas sample volume"*"
 of 120 liters either intermittently (equal
samples, equally spaced) or continuously
over 3 hours.
  After collecting the sample, disconnect the
probe and,tubing from the SOi scrubber and
 allow to cool. Before conducting the next run,
do the following. Clean the inside surface of
 the probe using a nylon brush and deionized
 distilled water from a wash bottle.until  the
rinse shows no visible particles. Replace the
probe filter. Thoroughly rinse the sample line
 connecting the probe to the scrubber until all
 visible particles are removed.
  4,2  Sample Recovery. Disconnect the
 implngcra. Replace the SO, scrubber contents
and the glass wool if saturated with solution
 for subsequent runs. Pour the contents of the-
midget impingers of the Method 6 part of the
train into a leak-free polyethylene, bottle for
shipment. Rinse the three midget impingers,
the connecting tubes, and the sample line
between the furnace and the first impinger
with deionized distilled water, and add the
washings to the same storage container.
Mark the fluid level. Seal and identify the
sample container.
  4.3  Sample Analysis. Note level of liquid
in container, and confirm whether any
sample was lost during shipment; note this on
analytical data sheet. If a noticeable amount
of leakage has occurred, either void the
sample or use methods, subject to the
approval of the Administrator, to correct the
final results.
  Transfer the contents of the storage
container to a 100-ml graduated cylinder.
Rinse the container with deionized distilled
water and add to the cylinder. Measure the
volume, and pour into a 250-ml Erlenmeyer
flask. Using the cylinder, add sufficient 100
percent isopropanol to give a final sample
concentration of 80 percent (v/v) isopropanol.
Add four to six drops of thorin indicator and
titrate to a pink end point using 0.0100N
barium perchlorate. Run a blank with each
series of samples.
  Note.—Protect the 0.0100 N barium
perchlorate solution from evaporation at all
times.
  5. Calibration—5.1 Metering System,
Thermometers, Rotameters, Barometer, and
Barium Perchlorate Solution. Fellow the same
calibration procedure as in Method 6,
Sections 5.1 to 5.5, respectively.
  5.2  System Performance Check. Using HjS
cylinder gas and combustion gas (as specified
in Sections 3.1.4 and 3.1.5), generate a series
of samples in the suspected concentration
range of TRS in the stack. Using the set-up
shown in Figure 16A-2, take at least two 30-
minute samples to determine system
performance'efficiency. Use the cylinder
regulator to set the combustion gas rotameter
flow rate to the desired level. Adjust the H2S
regulator to .a slightly higher than desired
flow rate to ensure excess gas for the system.
With the pump valve completely closed, turn
on the pump and open the valve slowly until
a 2 liter/min flow rate (or other selected flow
rate) is obtained. Observe the pressure
control vessel while opening the valve and
during the sampling run to maintain an
excess flow. The samples must be
transported through the entire sampling
system hi the normal manner. Compare the
resulting measured concentration to the
known concentration' by subtracting the
corrected volume of combustion gas from the
corrected total sample volume and treating as
in Section 6.3. The sampling system is  .
considered acceptable when two consecutive
samples of calibration gas produce results
which do not vary by more than ±5 percent
from their mean, and this mean-value is
within ±15 percent of the known value.
BILLING CODE 6560-26-M
                                                                 C-16

-------
o
i
               PRESSURE

               CONTROL'

                VESSEL
        H2S
COMBUST

   GAS
                                                                                             TUBE FURNACE
                                           SOzSCRUBBER
                                                                                                                       £2

                                                                                                                       g
                                                                                                                       a.
                                                                                                                       9.

                                                                                               3 PERCENT HzOz
                                                                      PUMP.
                                                                                                                       H


                                                                                                                       3.
                                                                                                                       10
                                        Figure 16A-2.  Calibration System
                                                                                                                       3
                                                                                                                       o
                                                                                                                       •a
                                                                                                                       o

-------
                 Federal Register  / Vol.  46.  No.  117  / Thursday. June 18. 1981 /  Proposed Rules          31909
  Conduct this performance check before the
test to validate the test procedure, sampling
system and tester. In addition, field
validation samples shall be taken to monitor
tosses in the probe due to absorption by stack
components. Perform this test by collecting a
known HiS sample (in the applicable
concentration range) after each third field
•ample and before cleaning the probe.
Introduce the gas into the probe and collect
in the usual manner. The obtained
concentration shall be within ±15 percent of
the known value. Otherwise, void the
previous three samples or make corrections
by dividing the sample concentration by the
fraction of recovery if the losses are between
0-20 percent. Substitute a field audit sample
for ono known sample during the collection
period if available. Such audit samples are
usually available from the Quality Assurance
Division, Environmental Monitoring Systems
Laboratory, U.S. Environmental Prbtection
Agency, Research Triangle Park, North
Carolina 27711.  ,
  6. Calculations—Carry out calculations,
retaining at least one extra decimal figure
beyond that of the acquired data. Round off
figures after final calculation.
  8.1   Standard Dry Sample Gas Volume.
Using Equation 6-1 of Method 6. calculate the
dry samplo gas volume V^dm) at standard
conditions'.
  0,2  TRS Concentration as SOj. Calculate
tho TRS concentration in ppm as SOa by
using Equation 6-2 of Method 6, except use
KSO*-12.020fi/meq.
  7. Bibliography—7.1 Curtis, F. and G.D.
McAlfster. Development and Evaluation of an
Oxidation/Method 6 TRS Emission Sampling
Procedure. Emission Measurement Branch,
Emission Standards and Engineering
Division, OAQPS, Environmental Protection
Agency, Research Triangle Park, North
Carolina 27711. February 1980.
   7.2  Blosser, R.O., H.S. Oglesby, and A.K.
Jain. A Study of Alternate SOj Scrubber
Designs Used for TRS Monitoring. A Special
Report by the National Council of the Paper
Industry for Air and Stream Improvement,
Inc., New York. N.Y. July 1977.
   7.3  Gellma. I. A Laboratory and Field
Study of Reduced Sulfur Sampling and
Monitoring Systems. Atmospheric Quality
 Improvement Technical Bulletin No. 81.
 National Council of the Paper Industry for Air
 and Stream Improvement, Inc., New York,
 N.Y. October 1975.
   7.4  Annual Book of ASTM Standards.
 Part 31; VViiter. Atmospheric Analysis.
 American Society for Testing and Materials.
Philadelphia, Pennsylvania, 1974. pp. 40-42.

[FR Doc, 81-18096 Filed 6-17-81; 8:45 am)
BILLING CODE 6560-26-M
                                                           C-18

-------
      APPENDIX D



EPA REFERENCE METHOD 6
           D-l

-------

-------
Chapter I—Environmental Protection Agency
App. A
                     METHOD 6—DETERMINATION or SULFUR DIOX-
                       IDE EMISSIONS FROM STATIONARY SOURCES

                     1. Principle and Applicability
                       1.1  Principle. A gas sample is extracted
                     from the sampling point in the  stack. The
                     sulfuric acid mist (including sulfur trioxide)
                     and the sulfur dioxide are separated. The
                     sulfur dioxide fraction is measured by the
                     barium-thorin titration method.
                       1.2  Applicability. This method is applica-
                     ble for the determination of sulfur dioxide
                     emissions  from  stationary  sources.  The
                     minimum  detectable limit  of the method
                     has been determined' to .be 3.4  milligrams
                     (mg) of SCVm3 (2.12x10-' lb/ft3). Although
                     no upper limit has been established, tests
                     have shown that  concentrations  as high as
                     80,000 mg/m3 of SO2 can be collected effi-
                     ciently in two midget impingers, each con-
                     taining 15 milliliters of 3 percent hydrogen
                     peroxide, at a'rate of 1.01pm for 20 minutes.
                     Based on theoretical calculations, the upper
                     concentration limit in a 20-liter sample is
                     about 93,300 mg/m,.
                       Possible interferents are  free ammonia,
                     water-soluble cations, and fluorides. The ca-
                     tions  and fluorides  are removed by glass
                     wool filters and an isopropanol bubbler, and
                     hence do not affect the SO, analysis. When
                     samples are being taken from a  gas stream
                     with high concentrations of very  find metal-
                     lic fumes  (such as  in inlets to control de-
                     vices),  a high-efficiency glass  fiber  filter
                     must be used in place of the glass wool plug
                     (i.e.,  the one in the probe) to remove the
                     .cation interferents.
                       Free ammonia interferes by reacting with
                     SOa to form particulate sulfite and by react-
                     ing with the indicator. If free ammonia is
                     present (this can be determined by knowl-
                     edge of the process and noticing white par-
                     ticulate matter in the probe and isopropanol
                     bubbler),  alternative methods,  subject to
                     the approval of the Administrator, U.S. En-
                     vironmental  Protection  Agency,  are  re-
                     quired.
                                      D-3

-------
App. A
Title 40—Protection of Environment
                                  D-4

-------
Chapter I—Environmental Protection Agency
                                 App. A
2. Apparatus
  2.1  Sampling.  The  sampling train  is
shown in Figure 6-1, and component parts
are discussed  below.  The tester has the
option of substituting sampling equipment
described  in  Method  8  in  place  of the
midget impinger  equipment  of  Method  6.
However, the Method 8 train must be modi-
fied to include  a heated  filter between the
probe and isopropanol impinger, and the op-
eration of the  sampling train  and sample
analysis must be at the flow rates and solu-
tion volumes defined in Method 8.
  The  tester also has the option of deter-
mining SOj simultaneously with particulate
matter and moisture determinations by (1)
replacing the water in a Method 5 impinger
system with 3 percent peroxide solution,  or
(2) by replacing the  Method 5 water im-
pinger system with a Method 8 isopropanol-
filter-peroxide system. The analysis for SO,
must  be consistent with  the procedure  in
Method 8.
  2.1.1 Probe.  Borosilicate glass, or stain-
less steel (other materials of construction
may be used, subject to the approval of the
Administrator), approximately 6-mm inside
diameter, with  a heating system to prevent
water condensation and a filter (either in-
stack or heated outs tack) to remove particu-
late matter, including sulfuric acid mist. A
plug of glass wool is a satisfactory filter.
  2.1.2 Bubbler and Impingers. One midget
bubbler, with medium-coarse glass frit and
borosilicate or quartz glass wool packed  in
top (see Figure  6-1) to prevent sulfuric acid
mist carryover,  and three 30-ml midget im-
pingers. The bubbler and midget impingers
must  be connected in series  with leak-free
glass  connectors,  silicone grease  may   be
used, if necessary, to prevent leakage.
  At the option of the tester, a midget im-
pinger may be  used in place of the midget
bubbler.
  Other collection absorbers and flow rates
may be used, but are subject to the approval
of the Administrator. Also,  collection effi-
ciency must be  shown to be at least 99 per-
cent for each test run and  must be docu-
mented in  the  report. If the efficiency is
found  to  be acceptable after  a series  of
three tests, further documentation is not re-
quired. To conduct the efficiency test,  an
extra absorber must be added and analyzed
separately. This extra absorber must not
contain more than 1 percent of the total
SO,.
  2.1.3 Glass Wool. Borosilicate or quartz.
  2.1.4 Stopcock Grease. Acetone-insoluble,
heatstable silicone grease may be used, if
necessary.
  2.1.5 Temperature Gauge. Dial thermom-
eter, or equivalent, to measure temperature
of gas leaving impinger train to within 1° C
(2' F.)
  2.1.6 Drying  Tube.  Tube packed  with 6-
to 16-mesh indicating type silica  gel,  or
equivalent, to dry the  gas  sample and  to
protect the meter and pump. If the silica gel
has been used previously, dry at 175° C (350°
F) for 2 hours. New silica gel may be used as
received. Alternatively, other types of decis-
sants (equivalent or  better) may  be used,
subject to approval of the Administrator.
  2.1.7  Valve.  Needle  valve,  to  regulate
sample gas flow rate.
  2.1.8  Pump. Leak-free diaphragm pump,
or equivalent, to pull gas through the train.
Install  a small  surge  tank between the
pump and rate meter to eliminate the pulsa-
tion effect of the diaphragm pump  on the
rotameter.
  2.1.9.  Bate-Meter. Rotameter, or equiva-
lent,  capable of" measuring flow rate  to
within 2 percent of the selected flow rate of
about 1000 cc/min.
  2.1.10  Volume Meter. Dry gas meter, suf-
ficiently accurate to measure  the sample
volume within 2  percent, calibrated at the
selected  flow rate and conditions actually
encountered during sampling, and equipped
with a temperature gauge  (dial thermom-
eter, or  equivalent) capable of measuring
temperature to within 3° C (5.4° F).
  2.1.11  Barometer.  Mercury,  aneroid,  or
other barometer  capable of measuring at-
mospheric pressure to  within  2.5 mm  Hg
(0.1 in. Hg). In many cases, the barometric
reading may  be obtained from a nearby na-
tional weather service station, in which case
the station value (which is the  absolute
barometric pressure) shall be requested and
an adjustment for elevation differences be-
tween  the weather station and sampling
point shall be applied at a rate of minus 2.5
mm Hg (0.1 in. Hg) per 30 m (100 ft) eleva-
tion increase or vice versa for elevation de-
crease.
  2.112  Vacuum  Gauge and Rotameter. At
least 760 mm Hg  (30 in. Hg) gauge and 0-40
cc/min rotameter, to be used for leak check
of the sampling train.
  2.2  Sample Recovey.
  2.2.1  Wash bottles. Polyethylene or glass,
500'ml, two.
  2.2.2  Storage Bottles. Polyethylene,  100
ml, to  store impinger  samples (one  per
sample).
  2.3  Analysis.
  2.3.1  Pipettes.  Volumetric type, 5-ml, 20-
ml (one per sample), and 25-ml sizes.
  2.3.2  Volumetric  Flasks. 100-ml  size (one
per sample) and 1000 ml size.
  2.3.3  Burettes. 5- and 50-ml sizes.
  2.3.4  Erlenmeyer Flasks. 250 mi-size (one
for each sample, blank, and standard).
  2.3.5  Dropping Bottle. 125-ml size, to add
indicator.
  2.3.6  Graduated Cylinder. 100-ml size.
  2.3.7  Spectrophotometer. To measure ab-
sorbance at 352 nanometers.

3. Reagents

-------
 App. A
    Title 40—Protection of Environment
   Unless otherwise indicated, all reagents
 must conform  to  the specifications estab-
 lished by the Committee on Analytical Rea-
 gents of the American Chemical Society.
 Where such specifications are not available,
 use the best available grade.
   3.1  Sampling.
   3.1.1 Water.  Deionized,  distilled to con-
 form to ASTM specification D1193-74, Type
 3. At the option of the analyst, the KMnO.
 test for oxidizable organic matter may be
 omitted when high concentrations of organ-
 ic matter are not expected to be present.
   3.1.2 Isopropanol, 80 percent. Mix 80 ml
 of Isopropanol with 20 ml of deionized, dis-
 tilled water. Check each lot of Isopropanol
 for  peroxide impurities as follows: shake 10
 ml of Isopropanol with 10 ml of freshly pre-
 pared 10 percent potassium iodide solution.
 Prepare a blank by similarly treating 10 ml
 of distilled  water. After 1  minute,  read the
 absorbance  at 352 nanometers on a spectro-
 photometer.  If absorbance  exceeds  0.1,
 reject alcohol for use.
   Peroxides  may be removed from Isopro-
 panol by redistilling or by passage through
 a column of activated alumina; however,
 reagent grade Isopropanol with suitably low
 peroxide levels may be obtained  from com-
 mercial sources. Rejection of contaminated
 lots may, therefore, be a more efficient pro-
 cedure.
   3.1.3  Hydrogen  Peroxide,  3   Percent.
 Dilute 30 percent hydrogen peroxide 1:9 (v/
 v) with deionized,  distilled water (30 ml is
 needed per sample). Prepare fresh daily.
   3.1.4  Potassium  Iodide Solution, 10-Per-
 cent. Dissolve 10.0 grams  KI in deionized,
 distilled water and dilute to 100 ml. Prepare
 when needed.
  3.2 Sample Recovery.
  3.2.1  Water.  Deionized,  distilled,  as  in
 3.1.1.
  3.2.2  Isopropanol, 80 Percent. Mix  80 ml
 of Isopropanol with 20 ml of deionized, dis-
 tilled water.
  3.3 Analysis.
  3.3.1  Water. Deionized,  distilled,  as  in
 3.1.1.
  3.3.2  Isopropanol, 100 percent.
  3.3.3  Thorin Indicator. l-(o-arsonopheny-
 lazo)-2-naphthol-3,6-disulfonic  acid,   diso-
 dium salt, or equivalent. Dissolve 0.20 g in
 100 ml of deionized, distilled water.
  3.3.4  Barium Perchlorate Solution, 0.0100
N. Dissolve 1.95 g of barium perchlorate tri-
 hydrate [Ba(ClO«)r3H,O] in 200 ml distilled
water and dilute to 1 liter with isopropanol.
Alternatively, 1.22 g of [BaCl,-2H,O] may be
used instead of the perchlorate. Standardize
as in Section 5.5.
  3.3.5  Sulfuric Acid  Standard,  0.0100 N.
Purchase  or standardize   to  ±0.0002  N
against 0.0100 N NaOH which has previous-
ly been standardized against potassium acid
phthalate (primary  standard grade).
4. Procedure.
   4.1  Sampling.
   4.1.1 Preparation  of  collection  train.
 Measure  15 ml  of 80  percent Isopropanol
 into the midget bubbler and 15 ml of 3 per-
 cent hydrogen peroxide into each of the
 first two midget impingers. Leave the final
 midget impinger dry. Assemble the train as
 shown in Figure 6-1. Adjust probe heater to
 a temperature sufficient to prevent water
 condensation. Place crushed ice and water
 around the impingers.
   4.1.2 Leak-check procedure. A leak check
 prior to the sampling run is optional; how-
 ever, a leak check after the sampling run is
 mandatory. The leak-check procedure is as
 follows:
   Temporarily attach a suitable (e.g., 0-40
 cc/min) rotameter to the outlet of .the dry
 gas meter and place a  vacuum gauge at or
 near the  probe inlet. Plug the probe inlet,
 pull a vaccum of at least 250 mm Hg (10 in.
 Hg), and  note the flow rate as indicated by
 the rotameter. A leakage rate not in excess
 of 2 percent of the average sampling rate is
 acceptable.

   NOTE: Carefully release the  probe inlet
 plug before turning off the pump.
   It is suggested (not mandatory) that the
 pump  be leak-checked separately,  either
 prior to or after the sampling run. If done
 prior to the sampling run, the pump leak-
 check  shall precede the leak check of the
 sampling  train described immediately above;
 if done after the sampling run, the pump
 leak-check shall follow the train leak-check.
 To leak check the pump, proceed as follows:
 Disconnect the drying tube from the probe-
 impinger assembly. Place a vacuum gauge at
 the  inlet  to either the trying tube or the
 pump, pull a vacuum of 250 mm (10 in.) Hg,
 plug or pinch off the  outlet of the flow
 meter  and then turn off the  pump. The
 vacuum should remain stable for at least 30
 seconds.
  Other leak-check procedures may be used,
 subject to the approval of the Adminstrator,
 U.S. Environmental Protection Agency.
  4.1.3 Sample collection. Record  the  ini-
 tial dry gas meter reading and barometric
 pressure. To begin sampling, position the tip
 of the  probe at the sampling point, connect '
 the  probe to the bubbler, and start  the
 pump.  Adjust the sample flow to a constant
 rate of approximately 1.0 liter/min as indi-
 cated by the rotameter. Maintain this con-
 stant rate (±10 percent) during the entire
sampling run. Take readings (dry gas meter,
tempertures  at dry gas meter  and  at  im-
pinger outlet and rate meter) at least every
5 minutes. Add more ice during the run to
keep the  temperture  of the gases leaving
the last impinger at 20° C (68° F) or less. At
the conclusion of each run,  turn  off  the
pump,  remove  probe  from the stack, and
record  the final  readings.  Conduct  a leak
                                        D-6

-------
 Chapter I—Environmental Protection Agency
                                 App. A
 check as in Section 4.1.2 (This leak check is
 mandatory.) If a leak is found, void the test
 run, or use procedures acceptable to the Ad-
 ministrator.to adjust the sample volume for
 the leakage. Drain the ice bath, and purge
 the remaining part of the train by drawing
 clean ambient air through the system for 15
 minutes at the sampling rate.
  Clean  ambient air can be provided by
 passing air  through a charcoal filter  or
 through  an  extra midget  impinger with  15
 ml of 3 percent H,Oj. The tester may opt to
 simply use  ambient air,  without purifica-
 tion.
  4.2  Sample Recovery. Disconnect the im-
 pingers after purging. Discard the contents
 of the midget bubbler. Pour  the contents of
 the midget impingers into a leak-free poly-
 ethylene  bottle  for shipment. Rinse the
 three midget impingers and  the connecting
 tubes with  deionized,  distilled water, and
 add the washings to the same storage con-
 tainer. Mark the fluid level. Seal and identi-
 fy-the sample container.
  4.3  Sample Analysis. Note level of liquid
 in container, and  confirm  whether  any
 sample was lost during shipment; note this
 on analytical data sheet. If a noticeable
 amount of leakage has occurred, either void
 the sample or use methods, subject to the
 approval of the  Administrator, to  correct
 the final results.
  Transfer the contents of the  storage con-
 tainer  to a  100-ml  volumetric flask  and
dilute to exactly 100 ml with deionized, dis-
tilled water. Pipette  a 20-ml aliquot of this
solution into a 250-ml Erlenmeyer flask, add
80 ml of  100 percent  isopropanol and two to
four drops of thorin indicator, and titrate to
a pink  endpoint using  0.0100 N  barium
perchlorate. Repeat  and  average the  titra-
 tion volumes. Run a blank with each series
of samples. Replicate Mirations must agree
within 1  percent or 0.2  ml, whichever is
larger.

  NOTE: Protect the 0.0100 N barium perch-
lorate  solution  from  evaporation at  all
times.
5. Calibration
  5.1  Metering System.
  5.1.1  Initial Calibration. Before its initial
use in the field, first leak check the meter-
ing system (drying tube, needle valve, pump,
rotameter, and dry  gas meter) as follows:
place a vacuum gauge at the  inlet to the
drying tube and pull a vaccum of 250 mm
(10 in.) Hg;  plug  or pinch off the outlet of
the flow meter, and then turn off the pump.
The vaccum shall remain stable for at least
30 seconds.  Carefully  release  the  vaccum
gauge before releasing the flow meter end.
  Next,  calibrate the metering system (at
the sampling flow  rate  specified  by the
method)  as follows: connect  an  appropriate-
ly sized wet test meter (e.g.,  1 liter per revo-
lution) to the inlet of the drying tube. Make
three independent calibration runs, using at
least five revolutions of the dry gas meter
per run. Calculate the calibration  factor,  Y
(wet test meter calibration volume divided
by the dry gas meter volume, both volumes
adjusted to the same reference temperature
and pressure), for each run, and average the
results. If any  Y value deviates  by more
than 2 percent from the average, the meter-
ing system is unacceptable for use.  Other-
wise,  use the  average as the  calibration
factor for subsequent test runs.
  5.1.2  Post-Test Calibration Check. After
each field test series, conduct a calibration
check as in Section 5.1.1 above, except for
the following variations: (a) the leak check
is not to be  conducted, (b) three, or more
revolutions of the dry gas meter may be
used,  and (c) only  two independent  runs
need be.made. If the calibration factor does
not deviate by more than 5 percent from
the initial calibration factor (determined in
Section 5.1.1), then the dry gas meter  vol-
umes  obtained during the test series  are ac-
ceptable.  If the calibration factor deviates
by more than 5 percent, recalibrate the me-
tering system as in Section  5.1.1, and for the
calculations,  use the calibration factor (ini-
tial or recalibration) that  yields the lower
gas volume for each test run.
  5.2  Thermometers.   Calibrate   against
mercury-in-glass thermometers.
  5.3  Rotameter.  The rotameter  need not
be calibrated but should  be  cleaned  and
maintained  according to  the  manufactu-
turer's instruction.
  5.4  Barometer.  Calibrate against  a  mer-
cury barometer.
  5.5  Barium Perchlorate  Solution.  Stand-
ardize  the   barium  perchlorate  solution
against 25 ml of standard sulfuric acid to
which 100 ml of  100 percent isopropanol has
been added.
6. Calculations
  Carry out calculations, retaining at least
one extra decimal figure beyond that of the
acquired data. Round off figures after final
calculation.
  6.1  Nomenclature.
CK,=Concentration of sulfur dioxide,  dry
   basis corrected to standard conditions,
   mg/dscm (Ib/dscf).
N=Normality of barium perchlorate titrant,
   milliequivalents/ml.           -
P,»r=Barometric pressure at the exit orifice
   of the dry gas meter, mm Hg (in. Hg).
POA=Standard absolute  pressure,  760  mm
   Hg (29.92 in. Hg).
Tm= Average  dry gas meter absolute  tem-
   perature, °K (°R).
T,M=Standard absolute temperature, 293° K
   (528° R).
Vc—Volume of sample aliquot titrated, ml.
Vm=Dry gas volume as measured by the dry
   gas meter, dcm (dcf).
                                         D-7

-------
App. A

 V»c,u)-Dry gas volume measured by the dry
    gas meter, corrected to standard condi-
    tions, dscm (dscf).
 Vwta-Total volume of solution in which the
    sulfur-dioxide sample is contained, 100
    ml.
 Vi— Volume  of barium perchlorate titrant
    used for the sample, ml (average or rep-
    licate titrations).
 V<>*= Volume of barium  perchlorate titrant
    used for the blank, ml.
 y=Dry gas meter calibration factor.
 32.03 ^Equivalent weight of sulfur dioxide.
  6.2 . Dry sample gas volume, corrected to
standard conditions.
              r,eJ\ / °bmA	jr y '•m °hmr
              ~Tl) \J\TJ~K*Y  Tm

                              Equation 6-1
   Title 40—Protection of Environment

  4. Patton, W. F. and J. A. Brink, Jr. New
Equipment and Techniques for Sampling
Chemical  Process  Oases.  J. Air Pollution
Control Association. 13:162.1963.
  5. Rom,  J. J. Maintenance, Calibration,
and  Operation  of  Isokinetic  Source-sam-
pling  Equipment. Office of Air Programs,
Environmental   Protection  Agency.   Re-
search  Triangle  Park, N.C.  APTD-0576.
March 1972.
  6. Hamil, H.  F. and D.  E. Camann. Col-
laborative Study of Method for the Deter-
mination of Sulfur Dioxide Emissions from
Stationary Sources (Fossil-Fuel Fired Steam
Generators).    Environmental   Protection
Agency, Research Triangle Park, N.C. EPA-
650/4-74-024. December 1973.
  7. Annual Book of ASTM Standards. Part
31; Water, Atmospheric Analysis. American
Society for Testing and Materials. Philadel-
phia, Pa. 1974. pp. 40-42.
  8. Knoll, J. E. and M. R. Midgett. The Ap-
plication of EPA Method  6 to High Sulfur
Dioxide   Concentrations.  Environmental
Protection Agency. Research Triangle Park,
N.C. EPA-600/4-76-038. July 1976.
where:
.Ki—0.3858' K/mm Hg for metric units.
   -17.64' R/in. Hg for English units.
  6.3 Sulfur dioxide concentration.
                            Equation 6-2
where:
.Ki-32.03 mg/meq. for metric units.
   =7.061 x!0"slb/meq. for English units.
7. Bibliography
  1. Atmospheric Emissions from Sulfuric
Acid Manufacturing Processes. U.S. DHEW,
PHS, Division  of  Air  Pollution.  Public
Health Service Publication No. 999-AP-13.
Cincinnati, Ohio. 1965.
  2. Corbett, P. F. The Determination of
SO, and SO, in Flue Gases. Journal of the
Institute of Fuel. 24:237-243,1961.
  3. Matty, R. E. and E. K. Diehl. Measuring
Flue-Gas SOi and SOS. Power. 101: 94-97.
November 1957.
  'Mention of trade names or specific prod-
ucts does not constltue endorsement by the
Environmental Protection Agency.
                                          D-8

-------
      APPENDIX E



EPA REFERENCE METHOD 9
           E-l

-------

-------
App. A
METHOD 8—VISUAL  DETERMINATION OF THE
  OPACITY OF EMISSIONS PROM STATIONARY
  SOURCES

  Many stationary sources discharge visible
emissions into the atmosphere; these emis-
sions are usually in the shape of  a plume.
This method involves the determination of
plume  opacity by Qualified  observers. The
   Title 40—Protection of Environment

method includes procedures for the training
and  certification of observers, and proce-
dures to be used in the field for determina-
tion of plume opacity.  The appearance of a
plume as viewed by an observer depends
upon a number of variables, some of which
may be controllable and some of which may
not be controllable  in the  field. Variables
which can be controlled  to an  extent to
which they no longer exert a significant in-
fluence  upon plume  appearance include:
Angle of the observer with respect to the
plume; angle of the observer with respect to
the sun; point of observation  of attached
and detached steam plume; and angle of the
observer with respect  to  a plume emitted
from a rectangular stack with a large length
to width ratio. The method includes specific
criteria applicable to these variables.
  Other variables which may not be control-
lable in the field are luminescence and color
contrast between the plume and  the back-
ground  against which the plume  is viewed.
These variables exert an influence upon the
appearance of a plume as viewed by an ob-
server, and can affect the ability  of the ob-
server to accurately assign opacity values to
the observed plume. Studies of the  theory
of plume  opacity and field studies  have
demonstrated that a plume is  most visible
and presents the greatest apparent opacity
when viewed  against  a  contrasting  back-
ground. It follows  from this,  and is  con-
firmed by field trials, that the  opacity of  a
plume,  viewed under  conditions  where  a
contrasting background is present can be as-
signed with the greatest degree of accuracy.
However, the potential for a positive error is
also  the greatest when a plume is viewed
under such contrasting conditions.  Under
conditions  presenting a  less  contrasting
background,  the  apparent opacity  of  a
plume is less and approaches  zero  as the
color and  luminescence  contrast decrease
toward zero. As a result, significant negative
bias  and negative errors can be made when
a plume is viewed under  less contrasting
conditions. A negative bias decreases rather
than  increases the  possibility that a plant
operator will be cited for a violation of opac-
ity standards due to observer error.
  Studies  have been undertaken to deter-
mine the magnitude of positive errors which
can be  made by qualified observers while
reading plumes under contrasting  condi-
tions and using the procedures set forth in
this  method. The results of these studies
(field trials) which involve a total of 769 sets
of 25 readings each are as follows:
  (1) For black plumes (133 sets at a smoke
generator), 100 percent of the sets were read
with a positive error*  of less than 7.5 per-
  •Por a set, positive error = average opac-
ity determined  by observers'  25 observa-
                                Continued
                                        E-3

-------
  Chapter I—Environmental Protection Agency
                                 App. A
  cent opacity; 99 percent were read with a
  positive error of less than 5 percent opacity.
   (2) For white plumes (170 sets at a smoke
  generator, 168 sets at  a coal-fired  power
  plant, 298 sets at a sulfuric acid plant), 99
  percent of the sets were read with a positive
  error of  less than 7.5 percent opacity; 95
  percent were read with a positive error of
  less than 5 percent opacity.
   The positive observational error associat-
  ed with an average of twenty-five readings is
  therefore established.  The accuracy  of the
  method must be taken  into account when
  determining possible violations of applicable
  opacity standards.
   1. Principle and applicability.
   1.1 Principle. The opacity of  emissions
 from stationary sources is determined visu-
 ally by a qualified observer.
   1.2 Applicability. This method is applica-
 ble for  the determination of the opacity of
 emissions from stationary  sources pursuant
 to § 60.1Kb) and for qualifying observers for
 visually determining opacity of emissions.
   2.  Procedures. The observer qualified in
 accordance with paragraph 3 of this method
 shall use  the following procedures for visu-
 ally determining the opacity of emissions:
   2.1  Position.  The qualified observer shall
 stand at a distance sufficient to provide a
 dear view of the emissions with the sun ori-
 ented In the 140° sector to his back. Consist-
 ent  with maintaining  the above  require-
 ment, the observer shall, as much as possi-
 ble, make his observations from a position
 such that his line of vision is approximately
 perpendicular to the plume direction,  and
 when observing opacity of emissions from
 rectangular outlets (e.g. roof monitors, open
 baghouses,  noncircular stacks),   approxi-
 mately  perpendicular to the longer axis of
 the  outlet. The  observer's  line  of sight
 should not Include more than one  plume at
 a time  when multiple stacks are  involved,
 and in any case the observer should make
 his observations with his line of sight per-
 pendicular to the longer axis of such a set of
 multiple stacks (e.g. stub stacks  on bag-
 houses).
  2.2  Field  records. The  observer  shall
 record the name of the plant, emission loca-
 tion, type facility, observer's name and af-
 filiation, and the date on a field data sheet
 (Figure  9-1). The time,  estimated  distance
 to the emission location, approximate wind
 direction, estimated wind speed, description
 of the sky condition (presence and color of
clouds), and plume background are recorded
 on a  field data sheet  at the time opacity
readings are initiated and completed.
  2.3  Observations.  Opacity observations
shall be made at the point of greatest opac-
ity in that portion of the plume where con-
tlons—average  opacity  determined  from
transmlssometer's 25 recordings.
 densed water vapor is not present. The ob-
 server shall not look continuously at the
 plume, but instead shall observe the plume
 momentarily at 15-second intervals.
  2.3.1 Attached  steam plumes. When  con-
 densed water vapor is present within the
 plume  as  it  emerges  from  the emission
 outlet,  opacity observations shall be made
 beyond the point in the plume at which con-
 densed water vapor is no longer visible. The
 observer shall record the approximate dis-
 tance from the emission outlet to the point
 in the plume at which the observations are
 made.
  2.3.2 Detached  steam plume. When water
 vapor in the plume condenses and becomes
 visible at a distinct distance from the emis-
 sion outlet, the opacity of emissions should
 be evaluated at the emission outlet prior to
 the  condensation of water vapor and the
 formation of the steam plume.
  2.4 Recording observations. Opacity obser-
 vations shall be  recorded to the nearest 5
 percent at 15-second intervals on an obser-
 vational record sheet. (See Figure 9-2 for an
 example.)  A minimum of  24  observations
 shall be recorded. Each momentary observa-
 tion recorded shall be deemed to represent
 the  average opacity of emissions for a 15-
 second period.
  2.5 Data Reduction. Opacity shall be de-
 termined as an average of 24 consecutive ob-
 servations  recorded at  15-second  intervals.
 Divide  the observations  recorded on  the
 record sheet into sets of 24 consecutive ob-
 servations. A set is composed of any 24  con-
 secutive observations. Sets need not be  con-
 secutive in time and in no case shall two sets
 overlap. For each set of 24 observations, cal-
 culate the  average by summing the opacity
 of the 24 observations and dividing this  sum
 by 24. If an applicable standard specifies an
 averaging time requiring  more than 24 ob-
 servations, calculate the average for all ob-
 servations  made  during the specified  time
 period.  Record the average opacity on  a
 record sheet. (See Figure 9-1 for an exam-
 ple.)
  3. Qualifications and testing.
  3.1 Certification requirements. To  receive
 certification as a- qualified observer, a candi-
 date must  be tested and demonstrate  the
 ability to assign opacity readings in 5  per-
 cent  increments  to  25  different  black
plumes and 25 different white plumes, with
an error not to exceed 15 percent opacity on
any one reading and an average error not to
exceed 7.5 percent opacity in each category.
Candidates shall be tested according to the
procedures  described  in  paragraph   3.2.
Smoke  generators  used  pursuant to para-
graph 3.2 shall be equipped with a smoke
meter  which  meets  the requirements of
paragraph 3.3.
  The  certification  shall  be valid for  a
period of 6 months, at which time the quali-
                                         E-4

-------
 App.*
   Title 40—'Protection of Environment
 flcation procedure must be repeated by any
 observer in order to retain certification.
  3.2 Certification procedure. The certifica-
 tion test consists of showing the candidate a
 complete run of SO plumes—25 black plumes
 and 25 white plumes—generated by a smoke
 generator. Plumes within each set of 25
 black and  25 white runs shall be presented
 in random order. The candidate assigns an
 opacity value to each plume and records his
 observation on a suitable  form. At the com-
 pletion of each run of 50 readings, the score
 of the candidate is determined. If a candi-
 date fails to qualify, the complete run of 50
 readings must be repeated  in  any retest.
 The smoke test may be administered as part
 of a smoke school or training program, and
 may be preceded by training or familiariza-
 tion runs  of the smoke  generator during
 which candidates are shown black and white
 plumes of known opacity.
  3.3  Smoke generator specifications. Any
 smoke generator used for the purposes of
 paragraph 3.2 shall  be  equipped with a
 smoke meter installed  to measure opacity
 across the  diameter of the smoke generator
 stack. The  smoke meter output shall display
 iiistack opacity based  upon  a  pathlength
 equal to the stack exit diameter, on a full 0
 to 100 percent  chart recorder scale. The
 smoke meter optical design  and  perform-
 ance shall meet the specifications shown in
 Table 9-1.  The  smoke meter shall be cali-
 brated as prescribed in paragraph 3.3.1 prior
to the conduct of each smoke reading test.
At the completion of each test, the zero and
span drift shall be checked and if the drift
exceeds ±1 percent opacity,,the condition
shall be corrected prior to conducting any
subsequent  test  runs.  The  smoke meter
shall be demonstrated, at the time of instal-
lation, to meet  the specifications listed in
Table 9-1.  This  demonstration shall be  re-
peated following any subsequent repair or
replacement of the photocell or associated
electronic circuitry including the chart  re-
corder or output meter, or every 6 months,
whichever occurs first.
    TABLE 9-1—SMOKE METER DESIGN AND
        PERFORMANCE SPECIFICATIONS
      Parameter
a. Light source..
b. Spectral response of
  photocell.
c. Angle oi view	
d. Angle of projection	.:
e. Calibration error	...
f. Zero and span
 drift±1% opacity, 30
 minutes.
g. Response time	
                          Specification
Incandescent lamp operated «l
 nominal rated voltage.
Photopic (daylight  spectral re-
 sponse of the human eye-
 reference 4.3).
15* maximum total angle.
15* maximum total angle.
±3% opacity, maximum.
                   5 seconds.
  3.3.1 Calibration. The smoke meter is cali-
brated after allowing a minimum of 30 min-
utes warmup by alternately producing  simu-
lated opacity of 0 percent and 100 percent.
When stable response at 0 percent or 100
percent is noted, the smoke meter is adjust-
ed to produce an output of 0 percent or 100
percent, as  appropriate.  This calibration
shall be repeated until stable 0 percent and
100 percent readings are produced without
adjustment. Simulated 0 percent  and 100
percent opacity values may be produced by
alternately switching the power to the light
source on and off while the smoke generator
is not producing smoke.
  3.3.2 Smoke meter evaluation. The smoke
meter design and performance  are to be
evaluated as follows:
  3.3.2.1 Light source. Verify from manufac-
turer's data and from voltage measurements
made at the  lamp,  as installed,  that the
lamp is operated within ±5 percent of the
nominal rated voltage.
  3.3.2.2  Spectral response   of  photocell.
Verify from  manufacturer's  data that the
photocell has'a photopic response; i.e., the
spectral sensitivity of the cell shall closely
approximate the standard spectral-luminos-
ity curve for photopic vision which is refer-
enced in (b) of Table 9-1.
                                        E-5

-------
Chapter I—Environmental Protection Agency
                                                   App. A
 1








C£
UJ
UJ

h~
i
i«




S 2 «


fe
i
g
ggii
«-• _j trt (_?
}_•-*(/)-i
       CO O  CD

       O O  O
£
     g; C5 in P s: =

     S S " a ? S
                                A
                                                               *
                                              S
                                                            i  i!
                                                            *  iS

                                                            I  II
                                                            Z  •>£
                                                            S  SI
                                                            fC  H-<
                                   £»••
                                   *
                              s  g«
                                            S-

                                                                   «

                                                                    '
                                                      •- P»J  5
                                                      4* C VI  V)
                 O  HJ •*"  **~  W  ^^


                 do         §



                    21
                                  E-6

-------
App. A
COMPANY
LOCATION
TEST NUMBET
DATE	
                           Title 40—Protection of Environment

FIGURE  9-2  OBSERVATION RECORD          TAGE     OF 	
                      OBSERVER
                      TYPE FACILITY 	
                      POINT OF EHISSIOUT
Hr.






























M1n.
0
1
?
T
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1<>
20
21
??
23
24
25
26
27
?R
29
Seconds
IT






























IB






























:•)<)





















j








45






























StEAM PLUME
(check 1f applicable)
Attached



























' ' :


Detached






























COMMENTS
















' »













  3.3.2.3 Angle of view. Check construction
 geometry to ensure that the total angle of
 view pf the smoke plume, as seen by  the
 photocell, does not  exceed 15°.  The total
 angle of view may be calculated:from: =2
 tan"'d/2L, where  6=total angle of view;
 d=the  sum of the photocell diameter+the
 diameter  of the  limiting aperture;  and
 L=the  distance from the photocell to  the
 limiting aperture.  The limiting aperture is
                         the point in the path between the photocell
                         and the smoke plume where the angle of
                         view is most restricted. In smoke generator
                         smoke  meters  this is  normally an orifice
                         plate.
                           3.3.2.4 Angle  of  projection.  Check con-
                         struction geometry  to ensure that the total
                         angle  of projection of the lamp on the
                         smoke plume does not exceed 15'. The total
                         angle of projection  may be calculated from:
                                       E-7

-------
Chapter I—Environmental Protection Agency
                                 App.A
«»2 tan" 'd/2L, where 0= total angle of pro-
jection; d=  the  sum of the length of the
limp filament + the diameter of the limit-
  Ing aperture; and L= the distance from the
  lamp to the limiting aperture.
                  FIGURE 9-2  OBSERVATION RECORD
                            (Continued)
 COMPANY	
 LOCATION   •
 TEST NUMBlT
 MTE	
                    PAGE
OF
OBSERVER       .
TYPE FACILITY
POINT OF EMISSTOTJT
Hr.






























M1n.
30
31
32
K
34
3S
36
'"'37
38
39
4rt
41
9
Seconds
ff






























1£>






























30






























4b






























	 STEAM PLUME 	
(check 1f applicable)
Attached






























Detached






























COWIENTS


























.



  3.3.2.5 Calibration  error.  Using neutral-
density filters of known opacity, check the
error between the actual response and the
theoretical linear response of the  smoke
meter. This check Is accomplished by first
calibrating the  smoke  meter according to
   3.3.1 and then inserting a series of three
   neutral-density filters of nominal opacity of
   20, 50, and 75 percent In the smoke meter
   pathlength. Filters  calibrated  within  ±2
   percent shall be used. Care should be taken
   when Inserting the filters to prevent stray
                                        E-8   •

-------
App.A
Title 40—Protection of Environment
                    light from affecting the meter. Make a total
                    of five nonconsecutive readings for each
                    filter. The maximum error on any one read-
                    ing shall be 3 percent opacity.
                      3.3.2.6 Zero and span drift. Determine the
                    zero and span drift by calibrating and oper-
                    ating  the smoke  generator in a  normal
                    manner over a 1-hour period. The  drift is
                    measured by checking the zero and span at
                    the end of this period.
                      3.3.2.7 Response time. Determine  the  re-
                    sponse time  by producing the series of five
                   .simulated 0 percent and 100 percent opacity
                    values and observing the time required to
                    reach stable response.  Opacity values of 0
                    percent and  100 percent  may be simulated
                    by alternately switching  the power to the
                    light source off and on while the smoke gen-
                    erator is not  operating.
                      4. References.
                      4.1 Air Pollution  Control District Rules
                    and Regulations, Los Angeles County Air
                    Pollution Control District, Regulation IV,
                    Prohibitions, Rule 50.
                      4.2 Weisburd, Melvin I., Field Operations
                    and Enforcement Manual for Air, U.S. Envi-
                    ronmental Protection Agency, Research Tri-
                    angle  Park, N.C., APTD-1100, August 1972,
                    pp. 4.1-4.36.
                      4.3 Condon, E.U., and Odishaw, H., Hand-
                    book  of Physics, McGraw-Hill Co., N.Y.,
                    N.Y., 1958, Table 3.1, p. 6-52.
                                     E-9

-------
                                     TECHNICAL REPORT DATA
                             (Please read Instructions on the reverse before completing)
1. REPORT NO.
  EPA-340/1-83-Q17
                                                              3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE


  Kraft Pulp Mill  Inspection Guide
              5. REPORT DATE
                 February 1983  (preparation)
              6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

 Ronald Hawks/Gary  Saunders
                                                              8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS

 PEDCo Environmental,  Inc. .
 505  S.  Duke St.,  Suite  503
 Durham, North Carolina   27701
                                                              10. PROGRAM ELEMENT NO.
              11. CONTRACT/GRANT NO.

                 68-01-6310'
                 Task No. 65
12. SPONSORING AGENCY NAME AND ADDRESS


  U.S.  Environmental  Protection Agency
  Stationary Source  Compliance Division
 Jfeshinoton. D.C.   20460	
              13. TYPE OF REPORT AND PERIOD COVERED
                 Final report
              14. SPONSORING AGENCY CODE
18. SUPPLEMENTARY NOTES
   EPA Project  Officer for this  report was Robert  Marshall, telephone:  (202) 382-2862
10, ABSTRACT
       This manual  presents technical  data on kraft  pulp mill processes  and control
  equipment design  and application.   The manual also includes inspection checklists
  for use by agency personnel in evaluating process  parameters and control  equipment
  operating conditions.  Major emphasis is placed on baseline analyses and  detection
  and elimination of operation- and  maintenance-related problems.
 7.
                                 KEY WORDS AND DOCUMENT ANALYSIS
                   DESCRIPTORS
                                                b.lDENTIFIERS/OPEN ENDED TERMS
                            c. COSATI Field/Group
  Operation  and maintenance
  Kraft pulp mills
  Particulate
  TR 5
  Recovery boilers
  Lime kilns
10. DISTRIBUTION STATEMENT

  Unlimited
19. SECURITY CLASS (TillsReport)
  Unclassified
21. NO. OF PAGES

  405
                                                20. SECURITY CLASS (Thispage)
                                                  Unclassified
                                                                            22. PRICE
EPA Fotm 2220-1 (R»v. 4-77)   PREVIOUS EDITION is OBSOLETE

-------

-------

-------

-------
 United States
 Environmental Protection
 Agency
Office of Air Quality Planning and Standards
Stationary Source Compliance Division
Washington D.C.  20460
Official Business
Penalty for Private Use  -
$300
                                           Publication No. EPA- 340/1-83-017
                                                                                                                              Postage and
                                                                                                                              Fees Paid
                                                                                                                              Environmental
                                                                                                                              Protection
                                                                                                                              Agency
                                                                                                                              EPA 335

                                                                          If your address is incorrocl. please change on (he above label;
                                                                          tear off; and return to the abova address
                                                                          If you do not dosiro toconttnua receiving this technical report
                                                                          series, CHECK HERE D; tsar olf label; and return it to the
                                                                          above address.


-------