United btates
Environmenta1 Proxection
Agency
Air Pollution Training Institute
MD20
Environment*! Research Center
Research Triangle Park NC 27711
DRAFT 2-1-83
Air
APTI
f\            Ol§^ A  P%  Introduction to Baseline
V/OUrSG  Ol-  i"^"0  Source Inspection Techniques
REFERENCE
 BASELINE TECHNIQUES FOR AIR
 POLLUTION CONTROL EQUIPMENT
   PERFORMANCE EVALUATION

-------
             FORusEwrm

           API! COURSE SI: 445

             DRAFT 2-1-83
     BASELINE TECHNIQUES FOR AIR
     POLLUTION CONTROL EQUIPMENT
        PERFORMANCE EVALUATION
                  by
           John R. Richards
         Engineering-Science
          501 Wlllard Street
           Durham, NC 27701
             919-682-9611
       Contract No. 68-01-6312
             Task No. 30
           Project No. 9430
           Project Officer
            Kirk E. Foster
Stationary Source Compliance Division
 U.S.  Environmental  Protection Agency
        Washington,  D.C.  20460
            February 1983

-------
                                DISCLAIMER


     This draft report was furnished to the U.S. Environmental  Protection
Agency by Engineering-Science, Inc., Durham, North Carolina 27701.  It is
not an official  policy and standards document.   The  opinions,  findings,
and conclusions are those of the authors and not necessarily those of the
Environmental Protection Agency.   Every  attempt  has been made to represent
the present state of the art as well as subject areas  still under evalua-
tion.  Any mention  of products or organizations  does  not  constitute  en-
dorsement by the United States Environmental Protection Agency.

     The U.S. Environmental Protection Agency does not endorse  any speci-
fic Inspection  procedure.   This  report  is  presented for  informational
purposes only.  Inspectors are encouraged to  seek information  concerning
specific procedures, especially safety procedures which have been adopted
at their agency or  company.   This report is not  intended  to  abbreviate,
alter or supercede these procedures.
                                    11

-------
                               ACKNOWLEDGEMENTS


     Engineering-Science appreciates the  assistance  provided by the EPA Task
Manager, Mr.  Kirk  E.  Foster.   Numerous  Engineering-Science  personnel  made
valuable contributions to this report.   Ms. Robin Segal! prepared the sections
concerning inspection  of  flares  and concerning the  legal  and administrative
aspects of plant inspection.  Mr. Cal Thames prepared the section on inspection
of carbon bed adsorbers.  Useful information was provided by Mr. Ted Michael is
and Mr. Macon Shepard concerning carbon bed adsorbers.
                                     ill

-------
                            TABLE OF CONTENTS

                                                                    Page

1.0   INTRODUCTION TO BASELINE INSPECTION TECHNIQUE	1-1

        1.1  The Baseline Technique	1-2
        1.2  Underlying Principles	1-3

2.0  ELECTROSTATIC PRECIPITATORS	2-1

        2.1  Components and Operation	2-1
        2.2  Inspection Procedures	2-8
        2.3  References	2-14

3.0  FABRIC FILTERS	3-1

        3.1  Components and Operation	3-1
             3.1.1  Pulse Jet Fabric Filters	3-1
             3.1.2  Reverse Air (Outslde-to-Inslde) Collectors	3-3
             3.1.3  Reverse Air (Inside-to-0uts1de) Collectors	3-4
             3.1.4  Shaker Collectors	3-5
        3.2  Inspection/Maintenance	3-5
             3.2.1  Pulse Jet Filters	3-5
             3.2.2  Reverse Air (Outslde-to-Inside) Collectors	3-7
             3.2.3  Reverse A1r (Inside-to-Outside) Collectors	3-8
             3.2.4  Shaker Collectors	3-9
        3.3  References	3-10

4.0  WET SCRUBBERS	4-1

        4.1  Components and Operation of Particulate Scrubbers	4-1
             4.1.1  Preformed Spray Scrubbers	4-2
             4.1.2  Tray-Type Scrubbers	4-3
             4.1.3  Packed-Bed Scrubbers	4-4
             4.1.4  Venturi and Ori fice Scrubbers	4-5
                    4.1.4.1  Venturi Scrubber	4-5
                    4.1.4.2  Variable Throat Venturi Scrubbers	4-5
                    4.1.4.3  Orifice Scrubber	4-5
             4.1.5  Mechanically Aided Scrubbers	4-5
        4.2  Inspection Procedures	4-6
             4.2.1  Opacity and Gas Flow Rate	4-6
             4.2.2  Static Pressure Drop	4-7
             4.2.3  External  System Inspection	4-11
                    4.2.3.1  Demister	4-11
                    4.2.3.2  Throat (Venturi Scrubbers)	4-11
                    4.2.3.3  Shell	4-12
                    4.2.3.4  Precooler	4-12
                    4.2.3.5  Sump, Recirculation Line, Make-up	4-12
                               Line and Purge Line
        4.3  References	4-14

-------
                             TABLE OF CONTENTS (Contd.)


                                                                     Page

 5.0  MECHANICAL COLLECTORS	5-1

      5.1  Components and Operation	5-1
           5.1.1  Simple Cyclones	5-1
           5.1.2- Multiple Cyclones	5-3
      5.2  Inspection Procedures	5-4
      5.3  References	5-6

 6.0  CARBON BED ADSORPTION SYSTEMS	6-1

      6.1  Components and Operation	6-1
      6.2  Inspectlon Procedures	6-4
      6.3  References	6-8

 7.0  FLARE	7-1

      7.1  Components and Operation	7-1
           7.1.1  Elevated Flares	7-2
           7.1.2  Ground Flares	7-7
           7.1.3  Forced Draft Flares	7-7
      7.2  Inspection Procedures	7-9
      7.3  References	7-11

 8.0  AUXILIARY EQUIPMENT	8-1

      8.1  VentH at1 on System	8-1
      8.2  Fan Evalnation	8-4
           8.2.1  Physical  Condition of Fan Housing	8-5
           8.2.2  Fan Operating Conditions	8-5
      8.3  References	8-8

 9.0  PROCESS EQUIPMENT	9-1

10.0  VISIBLE EMISSION EVALUATION	10-1

      10.1  Observation Procedures	10-1
      10.2  Basel 1 ne Eval uatlon	10-5
      10.3  Limits to the Use of Opacity	10-9
      10.4  References	10-9

11.0  SAFETY INFORMATION	11-1

      11.1  General Procedures	11-2
            11.1.1  Safety Training	11-3
            11.1.2  Medical Monitoring Program	11-3
            11.1.3  Written Safety Procedures	11-4
                                    vi

-------
                        TABLE OF CONTENTS (Contd.)

     11.2  Common Inspection/Maintenance Hazards	11-4
           11.2.1  Inhalation of Toxic Agents and Asphyxiants	11-5
           11.2.2  Accidental Falls	11-6
           11.2.3  Noise	11-12
           11.2.4  Head Protection	11-13
           11.2.5  Eye Protection	11-13
           11.2.6  Footwear	11-14
           11.2.7  Burns	11-14
           11.2.8  Heat Strees	11-14
           11.2.9  Cold Weather Conditions	11-15
           11.2.10 Electrical Shock	11-15
           11.2.11 Ionizing Radiation	11-15
     11.3  Confined Area Entry	11-17
           11.3.1  General Entry Procedures	11-18
                   11.3.1.1  Electrostatic Precipitators	11-19
                   11.3.1.2  Fabric Filters	11-20
     11.4  References	11-21

12.0  ADMINISTRATIVE AND LEGAL ASPECTS OF PLANT INSPECTIONS	12-1

     12.1  Introduction	12-1
     12.2  Preinspection Procedures	12-1
           12.2.1  File Review	12-1
           12.2.2  Inspection Announcement	12-3
           12.2.3  Inspection Equipment	12-3
           12.2.4  Plant Surroundings	12-4
           12.2.5  Visible Emission Ob-servations	12-5
           12.2.6  Plant Entry	12-4
                   12.2.6.1  Warrants	12-7
           12.2.7  Preinspection Interview	12-10
     12.3  Field Proceudres	12-10
     12.4  Post-Inspect ion Procedures	12-11
           12.4.1  Post-inspection Interview	12-11
           12.4.2  File Update and Report Preparation	12-11
     12.5  Handling Confidential Business Information	12-12
           12.5.1  Defining Confidential  Business Information	12-13
           12.5.2  Suggested Handling Procedures	12-13
                   12.5.2.1  Receipt of Confidential  Business Data...12-13
                   12.5.2.2  Handling in the Office	12-14
                   12.5.2.3  Privileged Data and Report Preparation..12-14
                   12.5.2.4  Potentially Confidential Information....12-14
     12.6  Use of Photographic Documentation	12-15
     12.7  Reference	12-16

APPENDIX A WORKING FILES	A-l

APPENDIX B FLOW CHARTING TECHNIQUES	B-l

APPENDIX C VISIBLE EMISSION OBSERVATION FORM	C-l

APPENDIX D OSHA REGULATIONS	D-l

APPENDIX E NIOSH GUIDELINES	E-l

APPENDIX F BASIC STATISTICAL METHODS	F-l

                                   vii

-------
               1.0   INTRODUCTION TO BASELINE INSPECTION TECHNIQUE

     The Baseline  Inspection  Technique has  been developed  to  aid  both  the
 source  operators  and  regulatory agency inspectors to  routinely  evaluate  air
 pollution control  equipment performance.   Early diagnosis of emerging  oper-
 ating problems  is  essential  to the prevention  of  permanent  equipment  damage
 and the minimization of emissions.
     The fundamental  principle  of  the  technique  is  that  control  device
 performance diagnosis is most accurate when observed operating  conditions  are
 compared with  site-specific baseline  data.   The specific  "historical"  data
 implicitly takes into account the numerous subtle factors which  can influence
 emissions.  Baseline  assessments  avoid the errors potentially introduced  by
 extrapolation of published literature values to a given facility.
     Control device instruments and  field measurements are  sometimes subject
 to error; therefore, baseline diagnosis is based on sets  of data comparisons
 rather than reliance  on  just one  parameter.   Even  when  some of the data  is
 unavailable or  suspect in  quality,  it is  still  possible  to  reach  meaningful
 and accurate conclusions using the remainder of the data.
     The purpose  of the  Baseline Inspection  Technique  is  to rapidly  iden-
 tify significant  changes   in  performance   and  the possible  reasons for  the
 changes.  The  technique  does not  necessarily  provide definite  evidence  of
 noncompliance with  regulations,  nor  does  it  necessarily provide  a  specific
 list of repairs required.
     This manual  contains  chapters detailing  baseline inspection  techniques
 for most of the commonly encountered equipment for air pollution  control.   In
addition, it includes  chapters which discuss visible emission  observation,
auxil iary equipment  inspection,  process   equipment  evaluation,  inspection
 safety procedures,  and  the  administrative and  legal   aspects  of  regulatory
inspections.
                                      1-1

-------
     At this  point  1t  1s  suggested  that  persons  utilizing the  Baseline
 Inspection Technique  have   a  technical  background—preferably  engineering
 experience.  As  with  any  work  involving  equipment, care  should always  be
 exercised.  Formal  safety  training is  highly recommended  for this  activity
 and  for any  field  work  involving air  pollution  control  equipment.   No single
 technique will  satisfy  all   source  characteristics  and inspection  circum-
 stances; whenever  necessary  inspectors  and  source  operators  should modify
 the  procedures outlined  1n this manual.
 1.1  THE BASELINE TECHNIQUE
     In the  Baseline Inspection  Technique  the inspector  should  start at the
 point  of  gas exhaust and  proceed "counter"  to  the  gas  flow  through to the
 process equipment.   Proceeding counter  to  the gas flow (backward through the
 system) tends  to  minimize  inspection  time and  reporting requirements  and
 maximize the  amount of useful information  obtained.  More  specifically, the
 Information on effluents and control  equipment gained early in the inspection
 is not only  the  easiest  to  obtain, it  can  be used later either to narrow the
 scope  of the inspection or to terminate the inspection without completing the
 most time-consuming  part  of the   evaluation—namely, the  process  equipment.
     The Baseline  Inspection procedures are listed  below  1n the  suggested
 temporal sequence beginning at the stack, proceeding backward through the sys-
 tem, and ending with the process equipment.  For regulatory agency personnel,
 there  are  some  additional  preinspection and post-inspection  steps  required.
 These  are  listed  separately  from the  Baseline  procedures and  presented  in
 the  last chapter  of the  manual entitled "Administrative and Legal  Aspects  of
 Plant  Inspections".
 Baseline Inspection Procedures
     Observe the stack effluent
     Check the continuous monitor(s)
     Measure the fan parameters and evaluate physical condition
     Analyze the control  equipment performance and physical conditions
     Check the ventillation system performance and physical conditions
     Evaluate process operating conditions
     Check raw materials and/or fuels
Preinspection Steps (regulatory agency personnel only)
     Review the source files
     Schedule the inspection
     Check  the portable inspection equipment

                                     1-2

-------
     Observe the plant  surroundings prior-to entry
     Request entry to the plant
     Interview plant official (s)
Postinspection Steps (regulatory agency personnel only)
     Interview plant official (s)
     Update  source files
     Prepare report
     The Baseline  Inspection  approach  offers particular  advantage  to regula-
tory agency  personnel in that it is very consistent with  the regulations.  An
inspector must  confirm  compliance with a  set of  regulations  which  normally
consists of  the following:
     1.   Opacity levels.
     2.   Operation and maintenance of continuous monitors.
     3.   Operation and maintenance of control devices.
     4.   Recordkeeping requirements for control  devices.
     5.   Raw material  characteristics (fuel surfur regulations).
     6.   Process operation rates (specific operating permit limits).
By proceeding counter to the  gas flow, the inspector can  determine  compliance
with the  first  four  regulations rapidly.   At the  same time,  the  performance
of the collector and fan provides a good indication of changes in raw  material
characteristics or process load.  Therefore, in  many  cases it  is  possible to
abbreviate inspections when compliance with items 5 and 6 is highly probably.
This helps   agencies  better  utilize  scarce  inspection   time  and  minimize
disruption of source personnel  schedules caused by the inspection.
1.2  UNDERLYING PRINCIPLES
     The Baseline  Technique  is  based  on  five  basic principles.   The  first
of these  is that  the operating characteristics and performance of each con-
trol  system is initially assumed to be unique.  This  is  because there  are  a
myriad of process  variables  and  control  device design  factors,  any one  of
which can  singly  or collectively influence  performance.   Often  it  is dif-
ficult, if not impossible,  to determine  reasons  why apparently similar units
opperate quite differently.   To  ensure the  accuracy  of the performance  eval-
uation, the  Baseline  Techniques utilizes  a comparison of present  operating
conditions against a historical  baseline level  for that  unit.   Each  variable
which has shifted  significantly is  considered a "symptom"  of  possible  oper-
ation problems.  During the inspection a  number  of comparisons are made and

                                     1-3

-------
 the  symptoms are diagnosed.  Acquisition of site-specific baseline data is
 important to the success of the approach.
      Experience during  numerous  previous Inspections has  indicated that
 many on-site monitors for air  pollution  control devices are either inaccu-
 rate or have  failed  entirely.  Due  to this  very  common  problem, sole use
 of on-site  monitors  1s  unreliable.  A second  basic  principle  of Baseline
 Inspections 1s that  portable meters are used 1n Heu of permanent on-site
 meters whenever feasible.  The use of portable meters Includes basic qua-
 lity assuance  checks.   During the Inspection  both the portable  meter and
 permanent meter  values  are recorded.   This  provides  a  means to  confirm
 the  operational .status  of the onslte gauges.
      The basic type  of  equipment  necessary  to perform a Baseline Inspec-
 tion 1s  listed in Table 1-1 along  with an  approximate  estimate  of cost
 (1980 dollars).   Regulatory agencies do  not  need a  complete set for each
 Individual  and  may need  only  one set  for  each  local   office.  The most
 expensive Items  are  the  self-contained  breathing  apparatus  and  these
 are  required  only  1f confined area entry Into  environments  conceivably
 having oxygen  deficiencies or toxic  agents  1s  anticipated3.  The  rest
 of the  safety  equipment 1s quite economical  and  certainly each  inspector
 should be fully equipped.
      Principle 3 of  the Baseline  Technique  1s simply  that a^  much infor-
 mation as possible will be used to evaluate performance  so  that problems
 can  be  identified  at the  earliest  time possible.   Fan operating condi-
 tions, fan  physical  conditions,  and control  device  Internal  conditions3
 provide especially  useful  Indicators   of  developing   problems and  are
 therefore Included  within the scope  of Baseline Inspection  Technique.
      The Inspection data  1s organized 1n a  coherent  fashion as  it is ob-
 tained.  In this manner the Inspector can expand or abbreviate an inspec-
 tion based on an examination of the initial data.  And thus principle 4 is:
 Diagnosis of operation problems 1s done by evaluation of sets of symptoms
 characteristic of common problems.  Reliance on sets of symptoms means

^Internal  inspection  of  control  equipment 1s  not recommended  unless each
 individual  performing the  Inspection  has received  specific and detailed
 training in  safety  procedures.   See  Chapter 11.0  SAFETY  INFORMATION.
                                  1-4

-------

TABLE 1-1. RECOMMENDED EQUIPMENT FOR
FIELD INSPECTIONS
Item Approximate cost (1980 dollars)
*1.
*2.
*3.
*4.
*5.
*6.
*7.
*8.
*9.
10.
11.
*12.
*13.
*14.
15.
*16.
*17.
*18.
19.
20.
*21.
*22.
23.-
24.
*25.
*26.
27.
28.

Hard hat
Safety shoes
Safety glasses (side shields)
Ear protectors
Oust masks
Cannlster type respirator
Coveralls
61 oves
Tool belt
Ropes and harness
Pry bar
Flashlight
Tape measure
Ouct tape
Stopwatch
Grounding cable
Differential pressure gauges
Gauge magnets
Pitot tubes
Velometer
Thermocoupl es
Multimeter
Tachometer
pH paper
pH meter
Combustion gas analyzer (03 4 C02)
Oxygen meter (for confined area entry)
Self-contained rebreather
(minimum of 2 required)
*Ind1cates equipment required for most inspection
ment
should be taken when necessary.
$ 2 to 5 5
$ 25 to $ 40
$ 5 to $ 100
$ 1 to $ 10
$ 10/100 masks
$ 15 to $ 30
$ 20
$ 3 to $ 8
$ 30
$ 50
$ 5
$ 10 to $ 15
$ 7 to $ 10
$ 5
$ 20 to $ 150
$ 25 to $ 35
$ 50 to $ 75
$ 5
$ 100 to $ 150
$ 200 to $ 400
$ 150 to $ 300
$ 10 to X 20
$ 100 to $ 200
$ 3 to $ 6
$ 100 to $ 400
$ 150
$ 500 to $ 700
$1000 to $1500

situations. Other equip-

that analyses can  be  completed  even  if several  of the measurements  can-
not be made at a given facility.
     Principle 5 of the  Baseline Technique is that  the  judgement  of the
Inspector must take precedent over any procedures  contained  in  this re-
port.  Due to the  numerous  site-specific factors, an inspector must  al-
ways remain alert regarding necessary changes  in the  scope of the  inspec-
tion or procedures.  For example, nothing should be attempted at the risk
of personal injury  or damage to the  equipment.   Furthermore,  it  may  be
necessary to alter certain analyses.   Many inspectors will quickly  evolve
to more sophisticated  inspection techniques not  discussed herein.   Basic-
ally, it is impossiible to prepare a  concise,  reliable  inspection  manual
for all situations.  There is no substitute for  a  well-trained, conscien-
tious inspector.
                                   1-5

-------
                       2.0  ELECTROSTATIC PRECIPITATORS

      The  two  most  Important elements in the systematic inspection of an elec-
 trostatic precipitator  are power  input  data and  plume  opacity.   The  power
 data,  which is  easily read off the transformer - rectifier (T-R) sets, can be
 used  to  determine  if the particle charging and migration  is  adequate.   Both
 of these  are  basic to the proper  operation  of any precipitator.   It  is  not
 sufficient, however,  to  simply maintain  adequate  power levels.  Other  fac-
 tors  such as  partial bypass  (sneakage)  and reentrainment  may  lead  to  high
 mass  emission rates.  Such conditions are  detected by increased plume opacity.
 Therefore, power input and opacity are complimentary  diagnostic tools.   With
 these  performance  indicators  it  is possible to routinely  evaluate  operating
 conditions by comparing  present levels with  baseline  levels.  Shifts  off the
 baseline  indicate  a need for a detailed inspection  of precipitator components.
      A brief  discussion  of the components  common to most  large  electrostatic
 precipitators3  is  useful  in developing an effective inspection approach.   The
 basic  operating principles are introduced in the following section.
 2.1 COMPONENTS AND OPERATION
     The  gas  stream  generated in  the  process  equipment is delivered  to  the
 precipitator  at duct  velocities  of 50 to 75 ft/s.   The  gas  stream  must  be
 slowed in order to  provide time  to  adequately treat the effluent; accordingly,
 the entrance  to the unit  is typically  an  expansion  chamber, often termed  the
 "inlet nozzle."  Here the  gas velocity drops more  than an  order of magnitude
 down to 3 to  8  ft/s.  Turning vanes and  perforated plates are  often  used to
 improve gas distribution across the face of the precipitator.
aThis section concerns the standard negative corona precipitators.  For infor-
mation concerning the  positive  corona devices often  used  for  collection  of
organic aerosols, see The Air Pollution Engineering Manual, AP-40, 1973.
                                     2-1

-------
      The  participate  laden  gas stream enters a  number  of parallel  passages
 comprised by  large  collection plates  and a  discharge  electrode  assembly.
 There are many  varieties of the latter, Including the standard weighted wires
 assembly  to  the  recent  rigid  frame  and  rigid  electrode  designs.   Typical
 widths of the passages are  9 to 12  1n.   The  discharge  electrode  system 1s
 connected to  a  high  voltage  power  supply  which  converts 440 to 480  volt A.C.
 power to   15  to  70  kV D.C.   power.   Most new  power supplies  Include  power
 controllers to  optimize power  Input.
      There are  normally a number of power supplies  on a given electrostatic
 precipitator.   Each  of these feed a  separate  region  of  the precipitator,
 termed a  field.0   There  are  usually 2 to  10 fields  1n  a series, and 2  to 8
 fields in parallel  on a unit.   It  is important  for  the  agency  inspector to
 know the  layout of fields and the  field numbering  system used at the plant.
      The  particulate  matter   passing  through  the energized electrostatic
 precipitator  is charged and then  1t  migrates to the collection plate at a rate
 dependent on  the  strength of the electrostatic field and the  number  of ions
 on  the particle (which depends on  particle size).   Some particles also  are
 collected on  the  negatively   charged  discharge  electrodes.   Both  sets  of
 electrodes must be rapped on  a regular basis to remove most of the accumulated
 material.
      The  sol Ids dislodged during rapping  fall  into hoppers  below the  elec-
 trodes.   Most units have a pyramidal hopper.   Such  hoppers are usually equip-
 ped  with  insulation,  heaters  and level  indicators.   Beneath each hopper is a
 solids  discharge  valve which  could be  as simple as a  rotary  valve or  as
 sophisticated as a pressurized evacuation  or a  pneumatic evacuation  system.
     A  basic understanding of precipitator operating principles is necessary
 for the evaluation of the inspection results.  Figure 2-1 1s  a simple schematic
of the  particle charging process.   Beginning  in block  a, the  application of
the high  negative  voltage  to  the  discharge  electrodes  results  in  the  very
rapid acceleration of  electrons  away  from the  discharge electrode and toward
the collection plate.  The electrons are moving fast enough  so that  when  they
   field may include up to two separate bus sections.

                                     2-2

-------
  e
       Trajectory of Particle
                     with
                  Positive Ions
 Dust on   Electron
Discharge
 Electron
               Trajectory of
                  Negative
                    Ions
G'



O
      Discharge Electrode
      Collection Electrode
Gas Molecule

Positive Ion

Negative Ion

Electron

Dust Particle
                  ,  Dust Layer
                  O     on
                  I  Collection)
                      Plate
         Trajectory
             of
       Particle with
       Attached Ions
              FIGURE 2-1.   Particle  Charging  and Migration
                                  2-3

-------
 collide with  a  gas  molecule, that molecule-1 s Ionized.  The original electron
 plus  the one freed  during  lonization  are then  accelerated again  under the
 influence  of  the  strong electric field.   This ionization process is repeated
 many  times 1n a very  short distance, resulting in a large number of electrons
 and positive  gas  ions.   These positive ions return to the discharge electrode
 carrying a small  number of particles along (see Figure 2-1, block b).  As the
 electrons  continue  to  move  away from  the discharge  electrode,  the electric
 field strength  decreases.  The slower moving electrons are then captured dur-
 ing the next  encounter with a  gas  molecule  (see Figure 2-1,  block  c).  The
 negative ion  created  then begins to move  along the  electric  field lines to-
 ward  the collection plates.  The negative ions are themselves captured on the
 particles  (Figure 2-1,  block d); the larger the particle the more  ions which
 can be attached on  the  surface.   Once  the particles  cross the  turbulent zone
 between the discharge and collection electrodes, they  build up in a layer; the
 charge is  dissipated through the layer and onto the grounded collection elec-
 trode.  The ease  by which the  charge is dissipated through the dust layer is
 inversely  related to the  resistivity.
      The voltage  applied to the  discharge electrode (commonly termed the "sec-
 ondary voltage")  is one  of  the  most  important  operating  variables.   If the
 voltage is below what is called  the  corona onset voltage (see Figure 2-2), the
 avalance process  described earlier  is  not initiated.  Under these conditions
 the precipitator  acts simply as a settling chamber and the particulate matter
 removal  is usually  less than 40%.   As the  voltage  goes up  above  corona onset
 levels,  smaller particles are charged more effectively and the electric field
 which  drives  charged particles toward the collection  plate is increased.  The
 voltage  on  the  discharge electrodes should be as high  as possible limited only
 by  excessive  sparking  between the  electrodes.   Most  electrostatic precipita-
 tors operate  between 15 and 50 kV.
     A second parameter of importance is the current passing from the discharge
electrodes  to the  collection  plates  (commonly called  the "secondary current").
The secondary current  and secondary voltage  for each field as read  off the
panels of the T-R set control cabinet are multiplied together to get the total
power input to the  field  as  shown  in Equation 2-1.   This is conveniently re-
corded as watts  (kV  times mA equal watts).  The total power to the  precipitator
is the sum of power levels in each  field, see Equation 2-2.

                                    2-4

-------
             750
             500
           i-
           3
           u
           § 250
SPARKING
                     CORONA ONSET
                            I
                           10          25          40
                          Secondary Voltage, Kilovolts
                      FIGURE 2-2.  Voltage-Current  Plot

2.1.1  Moderate and High Resistivity Conditions
     The total  amount of power per unit volume of  gas being treated is  often
a good indirect indication of the degree of particle collection.   The  greater
the power input,  the greater the  collection efficiency.  The general  relation-
ship between power input and penetration  (1 minus  the collection  efficiency)
is shown in Equation 2-3.
                            x S,
                               Eq.  2-1
                    where PI » power to field 1, watts
                          Sv * secondary voltage of field 1,  kilowatts
                          Sc » secondary current of field i,  milliamps
                                    2-5

-------
                                                              Eq. 2-2
 t .



where
                                                              Eq. 2-3
                             a fractional  penetration

                          K.J * constant

                          Pc » corona power, watts

                          V  « fl ow rate ACFM x 103


     The mass penetration power Input curve which  results  from Equation 2-3

is illustrated in Figure  2-3.   The Kj  constant is dependent  on  many factors


including the particle  composition, surface  characteristics, and  size dis-

tribution.  The magnitude  of K^ is lowest for difflcult-to-collect materials.
             0.8
          c
          o


          5  o-6
          s-
          4->
          0)

          0)
          a.


          *  0.4

          o

          *j




         ^ 0.2
                        200       400       600      800


                         Corona Power, Watts/103  ACFM
      FIGURE  2-3.   Penetration as a Function of Power Input  and


                                   2-6

-------
     There are certain practical  factors  which modify  the  shape  of the pene-
tration-power input  curve  in actual  installations.   The low gas  velocities
and long  retention  times result  in effective settling  of  large  particles.
Therefore, even with  a  zero  power level, the penetration  rarely exceeds  0.6
(40% collection efficiency).  In each precipitator, there is a small  quantity
of gas which successfully bypasses the treatment  zone.   Due to this untreated
gas the penetration  never reaches  zero.   There are also  emissions  due  to  re-
entrainment which are only  partially affected by power  input.  Due  to these
factors, a  hypothetical  curve  for  an  actual  Installation  would  appear  as
shown 1n  Figure  2-4.  The two bottom areas  represent  the  penetration  due to
sneakage and reentrainment.   In  a given unit these values  could be considerably
greater than shown.
        0.020
     c
     o
     +J
     (Q
        0-015
     0.
     It)
     o
        0.010
        0.005
                                                  \
                      K-0.5  \   Ki=(X3

                                   •
                                   %


                                      %

                                           \
                                   Kj=0.1  \
                                                                    t
                         200        400         600
                             Corona Power, Watts/103 ACFM
                                         800
     FIGURE  2-4.
Influence  of  Reentrainment  and  Sneakage  on  Penetration
                   2-7

-------
      Due to the  strong  Influence  of the Kj parameter on  the  position of the
 curve and the  possible  Influence  of non-power related emission modes,  it is
 not presently  possible  to  define  a universally applicable  penetration  power
 input curve.  These correlations must be developed for each  installation sep-
 arately.  This is the only  means  to  adequately account  for  the  K-j  parameter.
 In some plants the particle characteristics may  vary enough  to cause substan-
 tial variations in the value of K-f.
      The use of power-emission correlations plus the evaluation of secondary
 voltage and current data used  to  calculate the  power provide powerful  tools
 for the  source and  agency  Inspectors.   Unfortunately, not  all  precipitators
 have secondary voltage  meters  and a few do not even have  secondary  current
 meters.  In such  cases (often with older units), similar analysis  can  be done
 with the primary current and  primary  voltage  meters which  are  almost  always
 avail able3-  The  primary meters monitor the power  input  to the transformer of
 the power supply.  These meters are a  less direct Indication  of  electrical
 conditions in the electrode zone.
 2.1.2  Low Resistivity
      When the resistivity of the dust layer 1s low,  there is no  general  rela-
 tionship between  the power  input  and the penetration.   Particulate emissions
 are primarily influenced by  the  superficial  velocities  through various  por-
 tions  of the predpltator,  the rapping  intensities,  the rapping  frequency,
 and the precipitator aspect  ratio.
 2.2  INSPECTION PROCEDURES
     The inspection  begins  with the stack visible  emission observation.   Com-
 pliance  with  visible emission  requirements 1s determined and any  change in
 opacity  from typical levels  1s  noted.   The  presence  of regular "puffs" should
 be  noted and  timed.   Next,  the  data  on  the output consoles of  any transmis-
 someters and  gas  monitors  should be obtained.  Again, the presence  of spikes
 should be logged  and the Intervals between spikes estimated.   The  transmis-
 someter  1s  usually 1n  a more  favorable location  and is more  sensitive to
 spikes than the visible  emissions observer.   The  present  level  and trend in
opacity over the  last  eight  hours  should  be  checked.   Also,  the   inspection
should note the frequency and levels  of the automatic span and zero checks.

awhen using primary  side data, the power input  1s usually multiplied  by the
 Power factor.  For  older  installation  a  factor  of 0.75 is  often assumed.
                                     2-8

-------
     An oxygen content  measurement  1n the  ESP  exit breeching  is  useful   in
determining if there is a serious air Inleakage.   Temperature  readings  at  the
exit are  also  useful   for  confirming  the  magnitude   of  the  inleakage.
     These preliminary steps accomplished before  the actual  inspection  of  the
Precipitators take  about  1/2 to 1 hour.   It  is time well  spend since this
data can be used to narrow the scope  of the inspection.
     The inspector then proceeds to  the electrical  cabinets  which are connec-
ted to each of the transformer-rectifer (T-R) sets.   For each T-R set  there  are
a set of meters on the outside of the cabinet. The meters  generally have  the
following units:
     primary voltage -   volts, A.C.
     primary current -   amps, A.C.
     secondary voltage - kilovolts, D.C.
                         secondary  current  - milliamps, D.C,
                         spark  rate -  number per minute
The readings  should  be  written  down  in the same order as the power  supplies
are arranged on the precipitator.1*2  For example,  assume there  is  a  precipi-
tator with 8 power supplies arranged 1n the order shown  1n Figure 2-5.
                                   CHAMBER
     FIGURE 2-5,
               CHAMBER  B
Precipltator Having  Two  Chambers  Each  with Four  Fields
    1n Series and Two Bus Sections per Field
                                       2-9

-------
      For the  first  chamber the  Inspector would  record all  the  electrical
 parameters on fields 1 to  4  in  order.  Then the values  on the next  chamber
 would be  recorded  starting with 5 and  ending  with 8.  The  reason for  this
 approach is that the inspector must  determine  if the  normal  trends are  pre-
 sent in order to confirm the prevailing resistivity range.  In  the  large major-
 ity of  the  predpitators  which have  moderate  resistivity,  the  secondary
 current is  very low- 1n the-Inlet  fields  and  Increases progressively 1n  the
 subsequent fields.   The trend  line 1s  shown  1n Figure 2-6.   The secondary
 voltages are usually slightly higher 1n the Inlet  field and  decrease toward
 the outlet field.   This trend line  1s  shown 1n  Figure 2-7.
      When the resistivity  1s low, all  of the  fields have high secondary  cur-
 rents which approach the  current limit  of  the T-R set.   Sparking 1s very  low
 in all   fields and may  be  absent  entirely  from the  outlet  fields.   When  the
 dust resistivity is  high the secondary currents are low throughout the chamber,
 however, there may  be a slight Increase  going from Inlet to outlet.  Sparking
 is usually but not  always  high when the  resistivity  1s  high.
      With the electrical  data logged 1n the proper  order,  1t is possible to
 quickly scan the data.   Data analysis can be  done most effectively by plotting
 the secondary voltage and  current plots as shown  1n  Figures 2-6 and 2-7.  The
 line marked baseline is the data recorded earlier,  generally  during  a stack
 test.  This sort of simple  plot can  be used to answer one very basic but vital
 question, namely, has there been  a  shift 1n all  the values 1n  each field or
 have only a few fields changed?  The latter normally  implies some localized
 problem such as  rapper  failure, malalignment, solids buildup, Insulator leak-
 age,  and air Inleakage.  If all the fields have changed, then the resistivity
 has  probably changed.   Based on the analysis of these plots, the remainder of
 the  Inspection can be arranged.   If the total  power Input has  not changed sig-
 nificantly  since the  baseline period and the opacity has not Increased signi-
 ficantly, then there 1s probably no  need  for further Inspection at this time.
     Table  2-1 Illustrates some of the ways the  inspection approach  can be ad-
justed to  respond to possible signs of  performance  problems.   If there  is a
need to  check the rappers,  the  Inspector should use whatever hearing protec-
tion and respirators3 are appropriate.  The quickest way to check  rappers 1s

dFugitive gas  leaks  on  predpitators  operated  under  positive  pressure  can
 lead to high concentrations of 503.
                                      2-10

-------
to have an electrician  access  the  rapper control cabinet and manually activate
each set of  rappers.   The  more patient inspector may  simply  let  the  rappers
activate on  their normal  schedule and confirm that each  sounds normal.   For
large units, the latter approach can be quite tedious.  Another approach,  used
by a number of operators, is to precariously balance a washer (or penny) on the
side of each  rapper.   When returning  several  hours later,  one need  only  find
the rappers that failed to shake the washer off.
1200
1
Q.
•.•


000
800
600
400
200
1 » • *
,
Baseline Values ^^
^*^'°''
vr^ + + * Present Values
o-- -•'
i i i i
1234
Field
FIGURE 2-6.  Comparison of Present and Baseline Values for Secondary Current,
                                  Four Field Unit
FIGURE 2-7.
                  45

                  40
               £ 35
               o
               ^30
                 25
                 20
                                    Baseline Values
                                        — D
                                     Present Values
                1234
                         Field

Comparison of Present and Baseline Values for Secondary Voltage,
                   Four Field Unit
                                     2-11

-------
      The  Intensity  of  rapping  should  be  checked whenever  puffing  is  ob-

 served3.  For  compressed  air  type units,  the  intensity  is  proportional  to

 the air  pressure  (usually gauges  are  available on the  air lines leading  to

 the rappers).5   For other types  of  rappers, the  intensity is  controlled  by

 either the voltage  or current  in the rapper power  lines.


           TABLE 2-1.  INSPECTION ORIENTATION BASED ON INITIAL REVIEW
                             OF ELECTRICAL SET DATA
                 Symptoms
 Electricalset
Other
Inspection items
 Low secondary currents
   (one or more fields)

 Reduced power input in
   all fields
Increased opacity
Increased opacity
 High power input in all
  or most fields
Puffing
 Low secondary voltage in
  one or more fields
Increased opacity
 Reduced voltages in a
  number of fields
Higher 03 levels
Check collection plate
rappers

Check for changes which
could lead to high par-
ticle resistivity (fuel
sulfur content, temper-
ature change,  ash  com-
position change)

Check rapper intensity
and sequence, especially
on inlet and outlet
fields; also  check  for
factors which would give
low particle resistivity

Source personnel should
check insulators during
next outage; agency per-
sonnel confirm that in-
sulator heater and pent-
house blowers are
operating

Check for air inleakage
around hatch and hoppers
     For  most  precipitators  it  is difficult to  confirm quickly  that  solids

are being discharged.   If  this  can be checked  safely and  reasonably quickly,
it is  a  worthwhile task.  If no  solids  are being discharged over  a 10  to  30

minute period,  then either  a solids  discharge system  problem  has developed

(with potentially  serious  consequences)   or  the precipitator  is   in  a  very

advanced  state  of  deterioration.   In either  case  immediate  attention  is

warranted.

                                   2-12

-------
     There  is  occasionally  a  need to make an internal  inspection of at least
part of  the  precipitator  to  identify  an   operating  problem.   An  internal
inspection  is  usually done only  by the  source  personnel;  however,  in  rare
instances the  agency  inspector may also find one  useful3.   Such inspections
should be done in the company  of source personnel  completely  familiar  with
safety procedures  and only  during major  outages  when  the  units have  been
cooled, purged,  and deenergized.   Once  inside  the  region(s) exhibiting  un-
usual electrical  readings,  an  observer  would  check for any or all  of  the
following items:
     1.   Deposits  on collection  plates  (should be 1/8" to 1/4"  and dry),2
     2.   Wire  plate  alignment  (should be +_  1/4", visually should not be  able
          to  see  out-of-al ignment electrodes),2
     3.   Presence  of "clinkers"  at bottom of electrodes,
     4.   Presence  of broken wires,
     5.   Corrosion of electrodes, and
     6.   Eroded  or plugged gas distribution plates.
The importance  of conducting  the internal inspection  safely cannot  be over-
emphasized.   Personnel  should  not work  alone,  should  have  adequate  respira-
tors, should  use  grounding strips, and should follow all  lockout procedures.
     After  each  major outage,  source  personnel should  perform an  air  load
test on each field and record the results of the  test  in a convenient location.
This test will  clearly  identify uncorrected malalignment or the presence of
foreign objects  left  by  personnel  inside  the  unit.   Also  insulator  high
voltage leakage can be  identified.   The  agency  inspector  should request  such
records whenever these problems are suspected.
     The inspection of the precipitator  should  also  include  at  least a brief
review of the operating  records  kept by  the  source.   Such records  should
obviously include  the T-R  set electrical  data (hopefully  analyzed  by  the
operator, not just filed  away).  Also,  the  records should describe any signif-
icant maintenance  work  such as the  addition of wire  shrouds,  adjustment of
rappers, and the removal  of wires.  Concerning the latter item,  there should

^Internal  inspections are  not  recommended unless the  inspector has  received
 specific and detailed training in confined area entry.
                                    2-13

-------
be a drawing available  Indicating  where  wires  had been removed and  where on

any given wire  the  failure had occurred.  Based  on  the pattern  of  failures

an operator can determine 1f there 1s some  underlying problem leading to wire

failure, such as malallgnment.  Obviously  such problems should be corrected
before the wire 1s  replaced, otherwise the new wire will  also soon  fall.   With
such records both the source  personnel and  agency Inspector can  more  easily
determine what 1s reasonable and necessary to maintain continuous compliance.
2.3 REFERENCES

1.  Richards, J.,  Hawks, R.  and Szabo,  M.,  "ESP  Inspection  Procedures -
    Operator and Agency Roles in Ensuring Continuous Compliance",  In: Proceed-
    ings of Specialty Conference  on Operation and Maintenance of  Gas Cleaning
    Equipment, Air  Pollution  Control  Association,  April   1980,  pp  41-50.

2.  Ahrens,  J.,  "A  Coordinated Approach  by a  Central   Group  to Maintain
    Precipitator Compliance",  In:  Proceedings of  Specialty  Conference on
    Operation and  Maintenance  of  Gas  Cleaning  Equipment,  Air  Pollution
    Control  Association,  April 1980,  pp 51-60.

3.  Neundorfer,  M., " Electrode  Cleaning Systems:   Optimizing Rapping Energy
    and Rapping  Control", In:  Proceedings of  the  Symposium on the Transfer
    and Utilization of Particulate Control Technology, 1979.
                                    2-14

-------
                              3.0 FABRIC FILTERS

     In fabric filter  systems participate  is  collected  within  a dust  cake
supported on  either  a woven  or felted  fabric.   The  fundamental  collection
mechanisms include ^impaction, interaction,  diffusion,  and  in   some  cases
electrostatic attraction.  The latter has not been  fully  studied, therefore,
its importance on a  specific  unit  is difficult  to quantify.1  The collection
efficiency is extremely high  and any emissions  which  do  occur are thought  to
be due  to  leaks  through  bag  tears  across poorly seated  gaskets, cracks  in
welds, or high-velocity  pinholes.  As  a result, the typical  size efficiency
curve does not have a strong particle size dependence.2.3  p0r this reason, the
outlet particle size distribution is  often similar to the inlet distribution.
3.1  COMPONENTS AND OPERATION
     The manner by which  the  fabric  is  cleaned  and  the means  of  mounting the
bag in  the  housing   are  the  two principal features  used to  classify  fabric
filters.  For the purposes of these reports,  the four broad categories listed
below will be discussed separately.
          Pulse jet
          Reverse air (outside-to-inside flow)
          Reverse air (inside-to-outside flow)
          Shakers
Obviously there  are  a  number of  other types.   For  information  concerning
operating characteristics and components, see References  4 and 5.
3.1.1  Pulse Jet  Fabric Filters
     Example pulse jet filters are depicted in Figures 3-1 and 3-2.  The bags
are cylindrical  and are supported on an interior cage which is suspended from
the tube sheet.  The particulate laden gas stream is  admitted in the hopper
area.   The particulate  is collected  on the  outside  surface  of  the  felted
bags.   The cleaned  gas passes up  the  inside  of the bags and into the clean
air plenum above  the tube sheet.  Above each row of bags there is  a compressed
air supply line called  a  "blow tube."  The line has either a hole or nozzle
                                     3-1

-------
                 FIGURE 3-1.   Pulse Jet Filter
        (Courtesy of Environmental  Elements Corporation)
FIGURE 3-2.  Pulse Jet Filter;   Close-Up View of Air Manifold,
                      Blow Tube, and  Venturi
       (Courtesy of Environmental  Elements Corporation)

                             3-2

-------
directed downward toward each bag.  At regular intervals, a solenoid  activated
valve on each  tube  is opened momentarily, allowing  a  pulse  of air to  enter
each bag in  that row.  The  shock  wave dislodges the  cake.   By this means,
cleaning can be  done without taking  the  unit off-line.  The dislodged  dust
is collected within  the  hopper  and is discharged through a  rotary  valve  or
similar device.
     The pulse jet collectors normally  operate  at an air-to-cloth ratio  (gas
flow 1n acfm divided by available filtering cloth area)  of 4  to  8, well above
other collectors.   Felted  type  material   is  used  almost  exclusively.
3.1.2  Reverse Air (Outside-to-Inside Flow)
     These collectors tend to be single compartment,  smaller  units similar  to
the unit shown in Figure  3-3.   The bags are  supported on  an  inner cage which
is again suspended  from  an upper tube  sheet.   The  gas stream enters on the
lower side of the hopper area and the gas flow pattern  is  the same as  for the
pulse jet collectors discussed earlier.
        FIGURE 3-3.
Reverse  Air  Fabric  Filter  (outside-to-inside
     (Courtesy of the Carter Day Company)
               3-3
flow)

-------
     Cleaning of the bags is usually done.by  a  rotating  arm  located above the
tube sheet.  A  small  fan  pulls  filtered  air  for injection through  the  rota-
ting arm and down through the bag.
     These types of  collectors  usually operate at  air-to-cloth  ratios  of  1
to 4.  Unlike  pulse  jet  collectors,  woven  fabrics are  generally used.
3.1.3  Reverse Air (Inside-to-Outside  Collectors)
     The bags in this type of collector are mounted  in a  lower tube sheet and
suspended from the  roof  of the compartment.    Cages  are  not generally  used,
however, some bags  are manufactured with anticollapse rings to aid dust dis-
charge during cleaning.   Gas  flow  1s up through the  bags,  passing from the
inside to the outside.
     The bags are  mounted in two  or more compartments.   Periodically,  the
chambers are Isolated by  means of dampers and  filtered air is passed through
from the outside of all the bags  in  the compartment.
     Either a thimble or  snap ring assembly  is normally  used  to attach  the
bags to the tube sheet (Figure 3-4).   These  attachments  are a common source
                    Bag
                "*~ Thimble—*
                             Stainless
                             D-^-Steel
                                 Clamp
        Cell  Plate
            Thimble Connection
                Cell Plat
late-J[
 Cuff with
  Spring
"•"Steel"-1
Band Cell
                       Snap Band Connection
          FIGURE 3-4.
Bag-Cell Plate Attachments for Reverse Air
      (inside-to-outside) Baghouses
                                    3-4

-------
of operating  problems.   An  adjustable hook or cap  is  used  at the top of the
bag to provide proper bag tension, a  critical  factor.   These  types  of fabric
filters often  have  precleaners  and/or blast  plates  to reduce the  amount  of
large sized  abrasive particulate  reaching the  lower  portion  of the  bags.
Because of  the gentle  cleaning action  and  the type  of bag  mounting,  this
system can  accept  glass-fiber bags.   For this reason,  most  high  temperature
(>400°F) units are of this type.  Air-to-cloth ratios range from 0.5 to 3.0.
3.1.4  Shaker Collectors
     These collectors resemble the reverse air (inside-to-outside flow) units
with the exception that  fabric  cleaning  is done  by  a shaker  motor instead  of
reverse air.   Cleaning  is done  compartment-by-compartment.   Gas  flow  is  up
through the bags and outside to  the "clean side".  Typical air-to-cloth ratios
are 1.0 to 3.0.  Woven fabrics are used almost exclusively.
3.2  INSPECTION/MAINTENANCE
     The inspection of all fabric filters begins with the visible emission ob-
servation.  Of particular interest is  the presence of short-term spikes during
the cleaning  of  a  specific  compartment (reverse air units) or of  a  specific
row (pulse  jet  units).   Oxygen  content  of  flue  gas  downstream of  fabric
filters is  a  useful  indicator of ambient air  inleakage  problems  which could
lead to condensation  on  the  bag.   Pitot  traverse data  and/or  fan  data can  be
used to estimate the actual  air-to-cloth  ratio,  a  very  important  operating
variable.   Emissions are directly related to actual  air-to-cloth  ratio.6»?»8
3.2.1  Pulse Jet Filters
     Each of  the valves  (Figure 3-2;  which  supply  the blow  tubes  should  be
checked to confirm that  adequate  cleaning  of  each  row  of bags is  being done.
A firm "thud" should be audible if the valve is activated and is  operational.
The row being  cleaned  is easily identified by lightly  resting a  hand on the
valve.   Normally the  rows  are  cleaned  starting at  one end  and  proceeding
sequentially to the other end.   Rows  not being cleaned should be  marked and
recorded.   If this condition has persisted for some time, it  is probable that
a substantial dust  cake  has accumulated on the  outsides of  the  bags.   This
could lead to somewhat higher gas velocities and therefore higher particulate
emissions  through the remaining rows.
                                     3-5

-------
      If all  of the  valves  sound weak  or if  none  are  being activated,  then
 the compressed air  supply  should  be  checked.   The  pressure  gauge  on  the
 cleaning air manifold  is  usually in  the range  of  90  to  120  psig.a   Lower
 values may indicate inadequate cleaning.   In rare cases the  supply  line to a
 set of valves  is  inadvertently shut  off.
      The cleaning air supply  tank should have a small drain at the  bottom to
 periodically remove oil  and  water.   Also  the main  air  line  coming  to  the
 pulse jet filter  should  have a  dryer (type necessary depends partially  on
 climate).  Without provisions to remove water and oil, the bags can be "wetted"
 from inside leading again to  a higher  than anticipated static pressure.
      The presence of any short-term puffs with a duration  of 1 to  3  seconds
 is indicative  of small holes  or  tears  in the bags.   When the  row having  the
 damaged bag(s)  is  pulsed,  the dust  cake bridging  the hole  is  temporarily
 removed, leading  to a short-term high emission rate until the  hole is "healed"
 by a newly formed dust  cake.   Large  holes and tears often do  not  reseal, thus
 leading to a continuous emission.  The row causing short-term puffs  should be
 identified by determining which row is  being pulsed at the time of  the spike.
      The static pressure drop  across the  pulse jet collector should be  checked
 by removing*5 the  lines  to  the existing  monitor  and clearing  the  tap  hole.
 Portable gauges  supplied by  the  inspector  can  then be  used  to accurately
 measure the  static  pressure  drop.   High  values  indicate  higher  than normal
 gas flow rates,  partial  blinding  of the fabric, or inadequate  cleaning.   Low
 values  suggest  large tears  or serious  ambient  air  infiltration on  the  clean
 side.   The static pressure drop is not a very sensitive  indicator  of  bag  tears
 and holes.9  If a  detectable drop in static  pressure  drop has occurred,  it  is
 probable that  a substantial  fraction  of the  gas stream  passes  through  the
 holes rather than the filter cake.
     Severe  air inleakage is often due to poorly  seated access  hatches on  the
 top  of  the  collector.   Usually the air  inleakage  can  be  easily  heard  and
 felt when  the problem is severe.
aSome brands are designed to operate at a much lower pressure, 15 to 20  psig.
In these cases, the  amount  of  air per  square foot of bag surface is normally
greater than the conventional type pulse jet filter.
"Often plant  rules  require that this  is  done  only  by   plant  personnel.

                                     3-6

-------
     A pulse jet  collector  must be locked-  off  before  an internal inspection
can be done.  Once this is safely done, the clean side hatches can be removed
to inspect the blow tubes and the Venturis  (if present).  Clean side deposits
indicate poor  collection  efficiency.   During the  early stages  of equipment
deterioration, poor efficiency  can only be  identified  by  examining deposits
on the underside  of the blow tubes, directly above each bag.
3.2.2  Reverse Air (Outside-to-Inside Flow)
     The rate of  solids discharge from the collector should be checked by ob-
serving the amount of solids being discharged to a waste  bin  or being transpor-
ted along  a  screw conveyor.   Solids discharge  is  relatively  continuous with
this type  of collector, therefore, the rate of discharge needs to be observed
for 1 to 5 minutes.  Lower than normal rates can indicate any of the following:
     1.  Failure  of the reverse air fan,
     2.  Failure  of the internal cleaning arm,
     3.  Failure  of the rotary discharge valve below the hopper,
     4.  Solids bridging within the hopper,
     5.  Reduction in the quantity of process gas treated due to
         serious  air inleakage,
     6.  Substantial penetration through the bags, and
     7.  Low  particulate  matter   generation  rate   at  the  process  source.
All of these,  with the exception  of No.  7,  are serious  operating problems
which demand the  inspector's immediate attention.  The first three conditions
can be confirmed  quickly,  simply  by observing the equipment.   For combustion
sources,  air inleakage can be identified by measuring flue gas oxygen content
levels at various locations before and after the collector.  On other sources,
potential sites  for  air inleakage  such as  flanges, welds,  solid discharge
valves and hatches should be checked for audible air inleakage.
     The static pressure drop across the collector  should  be evaluated using
the same  procedure discussed earlier.   Higher than normal levels again suggest
inadequate cleaning or  blinding  of the outer bag surface  due to  moisture or
oil entering with the  gas stream.  Low static pressure drops  may be caused
by numerous tears or holes  in the woven  fabric or by poorly  attached bag
                                     3-7

-------
 cage assemblies at  the  tube  sheet.   As in  the  case with  pulse jet  units,
 there must be  numerous holes  and/or tears  before there is a  noticeable drop
 in static pressure.
      Internal inspection  should be done only when the unit (fan,  reverse air
 blower, cleaning arm and  rotary valve) is locked  off and all   safety  require-
 ments are satisfied.  Bag-to-bag  abrasion,  often caused by  bent cages,  can
 be identified  by  observing the bottom of the bags from the hopper  access
 hatch.3
 3.2.3  Reverse Air (Inside-to-Outside  Flow)
      The solids discharge from a typical multicompartment reverse air  fabric
 filter 1s Intermittent.   There is  significant solids discharge only  during  a
 short period while each of the compartments is undergoing cleaning.   The  dis-
 charge rate should  be checked during  one or more  cleaning  cycles.  Reasons
 for inadequate solids discharge are the same as  those  discussed earlier  for
 Reverse Air (outside-to-inside flow) collectors.
      Static  pressure drop across the entire collector should be measured  with
 portable gauges.   There should be  no need to  measure the pressure drop  across
 each compartment  since these must  be the same.   If there are  compartment-by-
 compartment  flow  resistance differences, the flow of  gas will automatically
 adjust to give equal  static  pressure drops.
      This type of collector is often  used  for high  temperature  service.  In
 such cases,  the  Inlet gas temperature  should be measured using a  portable
 thermocouple and then compared with the on-site instrument.  High  temperatures
 are of concern only  if the maximum rated temperature of the fabric being  used
 is  exceeded.   Low temperatures  Indicate the potential  for  condensation of
 volatile compounds and/or water on the interior bag surfaces.
      One problem often leading to  low  temperature levels is air inleakage  up-
 stream from  the collector.  On combustion  sources this  can  be  evaluated by
 measuring the  flue gas oxygen content.   Significant deviation  from normal  oxy-
 gen  levels combined with  low gas temperatures  may  indicate serious air inleak-
 age.
"CAUTION:  Opening  of a hopper access  hatch  must be done  with  extreme care
due to  the  potential  for hot, free  flowing  solids  behind the  hatch  which
could lead to serious burns and/or suffocation.
                                     3-8

-------
     Internal inspection can only be done when one or more of the compartments
are isolated.  Even then extreme care is warranted since most such collectors
have some leakage of gas across  the  isolation dampers.  Unless the compartment
is pressurized by  a  clean air  fan  or  the  entire collector  has   been  purged
out, it is recommended  that the inspector use a  self  contained  rebreather or
air-line respirator.3
     If the  clean  side  deposits are uniform and substantial  (>!"  to  2" deep
throughout)  it is  very  difficult-to determine the  source  or sources  of  the
problem.  Normally most bag problems in this type of collector occur near the
bottom, therefore, this area should be  checked first.  Common problems  include
bag abrasion on the opposite side from the inlet and cutting of bags on sharp
thimbles.  The latter may be aggravated by excessive tension which causes  the
fabric to rub against the sharp  edge of the thimble during  reverse air clean-
ing.  Low  bag tension  can  cause  creasing  of  the  bag  at  the bottom  which
results in abrasion and pinholes.
     Solids  buildup in  bags can  be checked by grabbing the  bag just above  the
thimble.  If the bag is  full  of solids, then no gas has been passing  upward.
If the  hopper below  is full,  gentle  massaging of the fabric  will   not  be
successful in cleaning the bag.
3.2.4  Shaker Collectors
     This type  of fabric  filter  is  entirely analogous to  the Reverse  Air
(inside-to-outside flow)  collectors  with  the  exception   of  the cleaning
mechanism.   Inspection  is done  in a similar fashion.
     Due to  the  more  strenuous cleaning  action,   bag  tension  is  even  more
important.   Excessive tension  can result in bag  cutting on the  thimbles  or
bags being yanked  free  of the thimble clamp.  These  types  of  fabric  filters
also suffer most bag problems near the bottom, close to the tube sheet.
     Leaks around  poorly  seated  snap  ring  bags  may also   reduce  collection
efficiencies.  Out-of-round rings  or poor  installation  can cause these prob-
lems.  The  snap  ring  type  of  bag attachment  is  also  vulnerable to  inlet
abrasion since there  is no  thimble to  protect the  bags  from the turning  gas
stream.
^Internal  inspections are  not  recommended unless the  inspector has received
specific and  detailed  traTnThg  in confined area  entry  safety  procedures.
                                     3-9

-------
3.3  REFERENCES

1.  Donovan, R. P., Turner, J. H.,  and Abbott,  J.  H.,  "Passive Electrostatic
    Effects in Fabric Filtration," In:   Second Symposium on the Transfer and
    Utilization of Participate Control  Technology,  vol.  i.ycontrol  of
    Emissions from Coal fired Boilers.  VerutittH; >•  P-»  Armstrong, ^.  A.,
    and Durham, M.editors, U.S. Environmental  Protection Agency Publication
    No. 600/9-80-039a, September 1980,  pp.  476-493.

2.  Turner, J.,  "Extending  Fabric Filter Capabilities,"  J.  of the Air Pollu-
    tion Control Association, Vol. 24,  No.  12, 1974.

3.  Ensor,  D.  S.,  R.  G. Hooper,  and  R.  W.  Schech,  "Determination  of  the
    Fractional Efficiency, Opacity Characteristics, Engineering  and  Economic
    Aspects of a Fabric Filter Operating  at a  Utility  Boiler," Electric Power
    Research Institute Report, EPRI  FP-297, November  1976.

4.  Billings,  C., and Wilder,  J.,  "Handbook of  Fabric Filter Technology,
    Vol. I," APTD 0690, NTIS PB200648,  GCA  Corp., Boston, 1970.

5.  Theodore,  L., and  Buonicore, A.  J.,  Industrial Air Pollution  Control
    Equipment for Particulates,  CRC Press,  Cleveland,  1976.

6.  Hall, R. R., R. Dennis,  and N. F. Surprenant,  "Fibers,  Fabrics,  Face Vel-
    ocity and Filtration," In:  A Specialty Conference on the User and Fabric
    Filtration Equipment  III. E.  R.  Fredrick,  editor,  APCA Specialty  Conference
    Proceedings, 1978, pp 156-169.

7.  Dennis, R. H.  A.  Klemm, "Verification  of  Projected  Filter  System Design
    and Operation," in Symposium on  the Transfer and Utilization of Particulate
    Control Technology. Vol. 2,  Fabric  Filters and  Current Trends in  Control
    Equipment. Venditti.  F.  P..  J.  A. Armstrong, and M.  Durham.  Editors. U.S.
    Environmental Protection Agency, Publication No.  600/7-79-044b,  February
    1979, pp.  143-160.

8.  Leith, D., M. W. First,  M. Ellenbecker, and D.  D.  Gibson, "Performance of
    a Pulse-jet Filter at High Filtration Velocities," In:   Symposium on the
    Transfer and Utilization of Particulate Control Technology:   Vol.  2.
    Fabric Filters and Current Trends in  Control  Equipment, Venditti.FT P.,
    J. A. Armstrong,  and M.  Durham, editors, U.S. Environmental Protection
    Agency, Publication No.  600/7-79-044b,  February 1979, pp. 11-26.

9.  Engineering-Science,  Inc., "An Investigation of a  Fabric Filter  Emissions
    Correlation," Final  Report to the  U.S. Environmental  Protection  Agency,
    Contract 68-01-4146,  Task 69, May 1980.
                                       3-10

-------
                              4.0 WET SCRUBBERS

     Wet scrubbers comprise a very large and diverse set of control devices,
all of which share similar particle collection mechanisms.  Inertia! impaction
and Brownian diffusion are the dominant collection mechanisms in most conven-
tional scrubbers.   Accordingly,  these  scrubber  systems  generally  exhibit
strong particle-size-dependent  performance.   Among  scrubber  types  there are
substantial differences  with  regard  to the  effectiveness  of the mechanisms,
the greatest  differences  occuring   in  the  particle  size  range  of 0.2  to
2.0 MinA-  Particles  in this  size range scatter light  effectively, therefore,
plume opacity  is  high  when the  particulate penetration in  this  size range is
high.  For  this  reason,   opacity  is  a  good  diagnostic tool.   Unfortunately,
it is not  always  easy to  evaluate the  actual  opacity  due  to condensed water
interference.
     A second indication of high  particulate  penetration is  a  decrease in  the
static pressure drop.   Usually  when  the pressure  drop is  high,  the inertia!
impaction conditions are optimized and the  penetration  in the  0.2  to 2.0 micron
range is  minimized.   Pressure  drop  can  be accurately  measured,  therefore,
it is a  very  useful  diagnostic  tool.  If either the opacity or pressure drop
has changed significantly since  the  baseline period,  a more detailed inspec-
tion is warranted.
4.1  COMPONENTS AND OPERATION OF PARTICULATE SCRUBBERS
                                 I
     Because of their diversity,  it  is  difficult  to  classify scrubbers  into
distinct generic  categories.   Scrubbers  that   are  grouped  on  the  basis  of
similar mechanisms may have greatly  varying geometric arrangements.  In this
discussion major  categories  of  scrubbers  are  arranged  in an approximate
ascending order of performance capabilities  and energy requirements.
                                     4-1

-------
     The  scrubber  liquid  may  be  used  for  particle collection  in  several
 distinct  ways.   The most  common is  generation  of  droplets,  which are  then
 intimately mixed with the gas stream.  Particle collection can also be achieved
 on  water  layers  or  sheets  surrounding some form  of packing material.   In  this
 case the  gas  stream is  forced to  negotiate an  intricate  path  around  the
 individual packing  elements.   The  high  velocity  passage   of  gas  may  also
 generate  "jets"  of  li,quid  in which  particle  collection may occur.
 4.1.1   Preformed Spray Scrubbers
     Preformed spray scrubbers require the least energy of the various scrub-
 bers and  they consequently allow the  highest  particulate  penetration,  especi-
 ally of small-diameter  particles.   Most preformed  spray  scrubbers  are highly
 efficient only for  particles larger than 10  umA diameter and are  relatively
 less efficient for  particles smaller  than 5 M^A  diameter.
     A  spray tower  is the  simplest  type of preformed spray scrubber,  consis-
 ting of a chamber  containing an array of spray  nozzles  (Figure 4-1).   Parti-
 culate-laden gases  pass  vertically up  through  the  tower  while  the  liquid
 droplets  fall by gravity countercurrent to the gas  flow.  Particle  collection
 is  primarily by  impaction.  While the  spray  scrubber  has low particle remo-
 val capability,  it  is  often useful for treating effluent gas streams  having
 high mass loadings  of large  diameter  particulate matter.
     The  cyclonic spray tower is similar to the  spray tower  scrubber  except
 that the  gas stream is  given a  spiral  motion.   In one typical configuration
 particulate-laden gases enter the  scrubber  vessel  tangentially at the  bottom
 and pass  upward  in  a  spiral  motion  around  a  centrally located array of spray
 nozzles.  Droplet migration is  crosscurrent  to the gas flow.   A  vane  type
 cyclonic  scrubber utilizes a system of vanes rather than a tangential inlet
to impart  cyclonic  motion.   Other  cyclonic scrubbers  consist  of  multiple
miniature cyclone tubes,  each  with a  separate  liquid  supply.   This  design
utilizes  a downward flow  pattern  for the  gases  in  contrast to the  typical
upward flow.   Cyclonic  spray towers generally operate with  a  static  pressure
drop of 4 to 8 in.  W.G.   Penetration  of particles  less than  2 wnA in  diameter
is typically quite high.*
                                     4-2

-------

.- __
A A
A A


*

A A A
A A A
I
r
                                                GAS  OUTLET

                                                DEMISTER

                                                     LIQUOR INLETS
                                                    GAS INLET
                                                 LIQUOR OUTLET
                         FIGURE 4-1.  Spray Tower.

4.1.2  Tray-Type Scrubbers
     A tray-type scrubber typically consists of a vertical tower with one or
more perforated plates  mounted  transversely inside.   In  such  a  scrubber the
liquid flows  from top  to  bottom  while the  gas  flows  from bottom  to  top.
Gases in the tray-type  scrubber mix  with the  liquid  as  it passes through the
openings in the plates.
     The perforated plates  of a tray-type  scrubber  are  often  equipped  with
impingement baffles or  bubble caps over the perforations forming impingement
surfaces.  As the  gas  passes upward  through  a perforation,  it  is  forced to
make a 180-degree  turn  into a layer  of  liquid.   The gas bubbles through the
liquid,  causing the particulate to impinge against  the  liquid  sheet.   The
                                     4-3

-------
 impingement  baffles  are  below the liquid level on the perforated plates  and
 for this   reason  are  continuously  washed  clean  of  collected   particles.
 Penetration  through  a  typical  impingement plate  is low  for particles  larger
 than  1 MmA2,  but  penetration of submlcrometer  particulate  is higher than  with
 some  higher-energy  scrubbers.   Pressure drop  through  a  typical  baffle plate
 is roughly 1.5  in. W.G.  per stage.  Addition of plates increases the scrubber
 pressure  drop,  but  does  not  proportionally  decrease  the   penetration  of
 submicrometer particulate.
 4.1.3 Packed-Bed Scrubbers
      In the  typical  packed-bed scrubber the liquid is introduced at the top
 of the  packed bed and trickles down through  1t;  the  packing  breaks down the
 liquid flow  into  a  film with a  large  surface area.  The  liquid  can also  be
 introduced concurrent  with gas  flow or crosscurrent  with  gas flow.  Packing
 materials  include  raschig  rings,  pall   rings,   berl  saddles,  tellerettes,
 intalox saddles,  and materials  such as  crushed  rock.  Packed beds  are  also
 constructed  with  metal  grids, rods,  or fiberous pads.  These scrubbers  are
 often used for  gas transfer  or  gas cooling, both of which are facilitated  by
 the large  liquid  surface area provided on the  packing.
      Plugging of a bed can  occur if the gas to be treated is too heavily laden
 with  solid particles.    Moving-bed  scrubbers  have  been developed  that  have
 less  propensity  for  plugging.  These  scrubbers  are  packed  with low-density
 plastic spheres,  which  are  free to  move  within the  packing  retainers.
      Packed-bed scrubbers are reported to have low penetration for  particle
 sizes down to 3 MmA  and can sometimes remove  a  significant fraction of  par-
ticulate in the range of 1 to 2 ^mA.  The standard countercurrent arrangement
requires the  greatest  liquid  flow  and  can   best  handle  heavier   loadings.
Crosscurrent  packed-bed  scrubbers  require much  less  liquid  flow, usually
operate at  lower  static  pressure drops,  and  rarely suffer  from  plugging.
Concurrent packed-bed  scrubbers  are  reportedly  more  efficient than  other
packed-bed scrubbers for the smaller particulates, but they typically operate
at higher  pressure drops.

                                     4-4

-------
4.1.4  Venturi and Orifice Scrubbers
     Venturi and orifice scrubbers are perhaps the most common particulate
removal devices, in part because they allow lower penetration of small parti-
cles than  most  other types  of  scrubbers.   These scrubbers  accomplish  supe-
rior particulate collection by generating small liquid droplets in the turbu-
lent zone  in  a  manner that creates a high  initial  relative  velocity  between
the droplets and the  particulate.   Inertial  impaction  capture of particulate
by the scrubbing  liquid is more efficient  in  these  highly  turbulent  proces-
ses, but  a price  is  paid  in  energy consumption to  achieve  the  low  penetra-
tion.
4.1.4.1  Venturi Scrubber - The simple venturi scrubber, often called a gas-
atomizing  spray  scrubber,  consists  of a  series  of  sprays  upstream from a
converging and diverging "throat" section  (Figure  4-2).  As the gas approaches
the venturi  throat,   the  velocity  and turbulence  increase.  The  high  gas
turbulence atomizes the liquid into  small  droplets  and increases interaction
between the droplets and the particulate.   Pressure  drops in  venturi  scrubbers
can range  from less than 4 to 100 in. W.G.
4.1.4.2  Variable Throat Venturi Scrubbers  -  Pressure  drop  and venturi  per-
formance are partially dependent on gas  velocity through the  venturi.   Several
variations of the standard  venturi  scrubber have  been developed  to  allow the
venturi throat  dimensions  to be  changed  as  gas flow rates  change.   Among
these scrubbers  are  the  plumb-bob  venturi,  the  flodded-disc  venturi,  the
moveable-blade venturi, the radial-flow venturi, and the variable-rod venturi.
4.1.4.3  Orifice Scrubber - In an orifice  scrubber, sometimes referred to as
an entrainment or self-induced  spray  scrubber, the  gas stream passes over a
pool of  scrubbing  liquid at high  velocity  just before entering  an  orifice.
The high velocity of the gas induced ("entrains") a spray of scrubbing liquid
droplets, which interact  with the  particulate  in and  immediately  after the
orifice.   Orifice scrubbers  have  moderate  pressure  drops (3  to  15  in.  W.G.)
and low  penetration  of particulate  2  to  3  MmA in  diameter  and  larger.
4.1.5  Mechanically Aided Scrubbers
     The mechanically aided  scrubbers utilize a mechanical  rotor or fan to
shear the  scrubbing  liquid into  dispersed  droplets.   These   scrubbers  use a
specially designed stator and rotor arrangement  which  produces very finely
                                     4-5

-------
     GAS INLET
     LIQUOR
     INLET
                                                         GAS OUTLET
                                                    LIQUOR  OUTLET
                        FIGURE 4-2.  Venturi Scrubber.
divided liquid droplets  that  are effective in the  capture  of fine particu-
late.  The low penetration of fine particulate,  however,  is  achieved at a high
energy cost.  Both wet fan and disintegrator type mechanically aided scrubbers
are subject to particulate  buildup or  erosion  at  the rotor  blades,  so they
are often preceded  by precleaning devices for  removing  coarse particulate.
4.2  INSPECTION PROCEDURES
4.2.1  Opacity and Gas Flow Rate
     The inspection of all particulate  wet  scrubbers  starts with an evalua-
tion of plume  opacity,  weather conditions permitting.   The residual  plume
                                     4-6

-------
 (after steam plume dissipation) has a direct relationship  to  the  mass  concen-
 tration in the effluent gas.  Procedures for evaluating wet  scrubber  opacity
 are discussed in Chapter 10.
      Gas flow rate through the wet  scrubber  should  be  estimated  by use of  a
 pltot tube or by analyses of process operation.   The flow rate is  an  important
 variable since 1t often determines the relative velocities between  the  parti-
 cles and the water (droplets or sheets) and therefore determines the  effective-
 ness of Inertial impaction.
 4.2.2  Static Pressure Drop
      Static pressure  measurements  should be made  before and  after  the wet
 scrubber.3  Often it  1s necessary  to clean  out the existing  static pressure
 taps prior to making  the measurement  since  these taps  are prone  to  plugging.
 At the  same  locations,  the gas temperature  should  be  measured.    (Note that
 water droplet Impaction on the temperature  probe may not allow measurement of
 a true dry bulb temperature.)
      Performance of  wet  scrubbers  can be  evaluated  by  comparisons  of the
 present "power input" (usually proportional  to  pressure drop)  to  the baseline
 level.  This  approach,  called  the  Contact Power  Method,   is  described  in
 references 3 and 4.
      The Contact Power approach is  based  on the concept  that  penetration is
 directly related  to  the  energy input  into  the  gas-liquid  contact.5'9  For
 gas-atomized scrubbers, the power  consumption is approximated  by  the equation

      PG = 0.158 AP                                                  Eq.  4-1
 where
      PQ * power consumed,  hp/1000  acfm.
      AP * gas phase pressure drop, 1n. W.G.
 When the  liquid stream adds  a  significant fraction  of the total  energy,
 Equation 4-2 is also used.   The total  energy is the  sum of the gas  and  liquid
 phase power consumption (Equation  4-3).
aln the case of  ventuM  scrubbers,  the downstream tap  should  be located  after
 the elbow  or  in  the  cyclonic  separator in  order to  avoid  boundary  layer
 separation phenomenon.
                                      4-7

-------
       PL  »  0.583  PfL                                                Eq*
 where
       Pf  »  fluid  pressure, pslg.
       L   »  Iiqu1d-to-gas ratio, gpm/1000 acfin.
       PT  «  PG  + PL                                                  E1«
 The scrubber collection efficiency 1s  then expressed as shown 1n Equation 4-4.
       NT  «  0P?                                                   Eq- 4~4
 where
                                       ( Inlet grain loading 1
      NT » number of transfer  units,  In (outlet grain loading  j, dimension!ess.
      PT * total power consumption, hp/1000 acfm.
     0,/3 * constants, dlmensionless.
 The constants  are parameters dependent on the characteristics of the particu-
 late matter.   Using this approach 1t 1s assumed that there are no independent
 effects due to throat velocity, I1qu1d-to-gas ratio, scrubber design and other
 parameters.
      In certain cases  it  1s  possible  to  achieve good  correlations using this
 approach.  Figure 4-3 shows the relationship between outlet loadings and sta-
 tic pressure drop for a set of scrubbers on coal-fired boilers.  Hesketh^ has
 developed an empirical equation relating to static pressure drop and penetra-
 tion.  As shown  1n  Figure 4-4 there  is  a good  relationship between  the two
 parameters.
      Pt - 3.47 AP-1-43                                              Eq.  4-5
 where
      P^ * penetration, dimension!ess.
    AP  « gas  phase pressure drop, 1 in. W.fi.
     The  contact power analysis Indicates whether sufficient energy is being
 expended.   It can  lead to erroneous conclusions 1f there has been a major shift
 1n  the inlet particle  size distribution.  Also, physical  problems  which can
 lead to excess emissions regardless of power input can also invalidate a con-
 tact power analysis.  Such  problems include poor gas-liquid distribution (i.e.
 venturi throat), particle regeneration, and  particle nucleation.  These can be
detected  only by Increased opacities.   The Contact Power Correlations also do
not appear  to be valid when there  is significant condensation of water vapor in
the scrubber or when the 11quid-to-gas ratio is below 5 gallons  per 1000 ACF.

                                     4-8

-------
   0.07


   0.06


   0.05


   0.04

<*_
o
•o
7:  o.o3
HJ
o
3   o.oo;
o
•4-J
HI
                                2              3         4

                 Theoretical  Power  Consumption, HP/10^ ACFM
    FIGURE 4-3.   Correlation  of  a  Coal-Fired  Boiler  Scrubber  Outlet
                   Dust  Loading with  Theoretical  Power  Consumptiony
                                   4-9

-------
     0.500
     0.200
     0.100
     0.050
    0.020
0)
0-
    0.010
    0.005
                   T	1    I   I   M  HI
                                •         '
                         i    i
r—i—i   i  i  i  IT
i
i
i
                        3   4  5  6 7 8  10        20   30

                          Static Pressure Drop, Inches W.G.
            50   70
    FIGURE 4-4.  Comparison of Predicted  Penetration as Calculated 1n
                             Equation  4-5 and Measured5
                                   4-10

-------
4.2.3  External System Inspection
     The components  of the  wet  scrubber  system  should  be  inspected on  a
regular basis.  This  usually does not  require  much time and  often  emerging
performance problems can be readily identified.
4.2.3.1  Demister  - Demister  problems are  usually  caused  by  plugging  or
scaling from  excessive material  buildup or mechanical failures.   One easily
Identifiable symptom is the  carryover and eventual fallout  of  droplets  from
the scrubber stack.  This is usually caused by reentrainment of the scrubbing
liquor as  it  passes through  openings in the  partially  blocked demister  at
higher than  design  velocities.   Another  sign   of demister  failure  is  the
buildup of material around the  lip of the  stack.   This carryover of material
can also buildup  on the  fan  blades and cause severe  vibrations  which should
be readily apparent.
     A properly  operating  demister washing  system should  prevent  excessive
material buildup from occurring.  The inspection should include an  investiga-
tion into  the  type and frequency of  demister washing  and the quality of  the
washing water.   If the  water has a  high  solids  content,   a cleaner  water
source may be  necessary.
     The most  accurate method  of determining  demister  performance  is  the
measurement of  the static  pressure   drop  across  the  demister.   The  normal
pressure drop  range is between 0.5 and 2.0 inches of water with most  applica-
tions less than 1.0  inches.  Any  measurement  outside  of this range indicates
significantly  decreased efficiency of the demister.
4.2.3.2  Throat (Venturi Scrubbers)-  The inspection  of the  throat  of  a  ven-
turi scrubber  should  include  an  evaluation  of the  ease  of access  to  the
liquor distribution  components.   If  the  nozzles  or  weirs  are not  easily
accessible, this  may  mean that  regular maintainence  is  very difficult  and
that maldistribution may occur.   This problem is particularly acute  when  the
nozzles are susceptible to abrasion and plugging.
     At the point of water  introduction  into the throat, a thermocouple should
be used to measure the  gas  inlet temperature.  A temperature  greater  than
300°F can be an indication of potential operating problems if the gas stream
                                     4-11

-------
 contains  high  organic or metallic  vapors.   Upon contact  with the liquor  in
 the throat these vapors will condense to form submicrometer particles.  Often
 the particles  form  after  passing the  zone of  optimum  inertia!   impaction,
 hence,  penetration of these particles is high.
     The  use  of  recirculated  liquor with a  high  solids content may also  lead
 to problems when the entering  gas stream 1s greater than 300°F.  Some of the
 atomized  liquor droplets may  evaporate  to dryness  releasing  a "regenerated"
 particle.  Thus, a   small  degree of  evaporation can  significantly increase
 mass emissions from  the scrubber.
     If the  venturi  has a  variable throat, the relative position  (open  or
 closed  since  baseline tests)  should be  noted.   This  position has  a  large
 effect  on  throat pressure  drop  since this  changes  the cross  sectional   area
 available  for the liquor and gas to pass through.  As the throat area decrea-
 ses, the  relative  droplet  particles  velocities  Increase, hence,  penetration
 is reduced.
 4.2.2.3  Shell - The exterior of the scrubber shell  is  inspected for corrosion
 and erosion.  Corrosion  is  quite probable  if the shell construction material
 is carbon  steel  and the pH  of the recirculated  liquor is less than  6.   The
 problems caused by erosion usually occur 1n areas of high velocity and direct
 contact of the  gas-liquor  stream.  The  most common areas of  damage  are the
 elbow of the venturi, opposite the tangential inlet  to  the cyclonic separator,
 and the inlet to the venturi throat.
 4.2.3.4 Precooler - The  presence of a  precooler should  be  noted during the
 inspection.  Certain types of  problems associated with the temperature of the
 gas stream as discussed  under the throat Inspection procedures above can be
 remedied through the use  of a  precooler.   The quality of the water  used in
 the precooler should be  determined.   Since most of  this  water evaporates,  a
 high solids content  can significantly increase  the mass loading  of  the  gas
 stream  that enters the throat.
4.2.3.5  Sump, Recirculation Line, Make-up Line  and Purge Line  -   An attempt
should be made to make  an  estimate of the  liquor  recirculation,  make-up and
purge rates if  at  all possible.  The most  suitable location  for  the recir-
culation and make-up rate  estimates  is  at  the  point  of discharge  into the
                                     4-12

-------
red rculation tank.   The  purge  rate  estimate  can  be made  at the  settling
pond or  lagoon.   If there  1s  some  apparent  or  suspected  problem with  the
scrubber, a  sample  of  the  recirculation  liquor can be obtained  from  the top
of the  recirculation  tank.   This sample can  be analyzed to determine  if it
has any  properties  which  might  adversely  affect the  system.   At this  time
the temperature  and pH of  the  recirculation  liquor can  also be measured.
In order to  document  the  recirculation system, a fl ow diagram of the  scrub-
ber's piping should be drawn.
     The static pressure drop  across the demister should be  measured.   Plug-
gage of  this can lead  to  reduced gas  flow through the wet scrubber  or to
carryover of  water droplets.   The  latter  can  result  1n   fan  Imbalance  or
local fallout.   Static pressure  drop  across  demisters  normally  range  from
0.5 to 3.0 inches W.6.
     Internal inspection  of  wet  scrubbers  1s normally  not  necessary  or
advisable.   It 1s usually  possible to check  for nozzle pluggage or  poor  liquor
spray patterns by  viewing the  nozzles  from an external  hatch  (when the  unit
has no  effluent  gas passing through and has  been  purged).  Other problems
can be  identified  equally well  simply from the  viewing ports of a  hatch3.
The entry  into a  wet  scrubber  should  be done  only by  plant  maintenance
personnel cognizant of potential damage which  can be done to corrosion  resis-
tant linings, trays and other components.
aHatches should be  opened  only  by plant personnel,  and all  safety procedures
 should be fully satisfied.
                                     4-13

-------
4.3  REFERENCES

1.   Theodore, L., and Buonlcore, A.  J.,  Industrial  A1r Pollution Control
     Equipment for Particulates. CRC  Press,  Cleveland 19/b.

2.   Calvert, S., J. Goldshmld,  D. Leith,  and D.  Mehta, Wet  Scrubber System
     Study. Volume I;  Scrubber  Handbook.  EPA-R2-72-118a,  August 1972.

3.   Engineering-Science, "Scrubber Emission Correlation," Final  Report to
     U.S. EPA, Contract No. 68-01-4146, Task Order 49,  1979.

4.   Semrau, K. et al., "Energy  Utilization  by Wet Scrubbers,"  EPA Report
     600/2-77-234, November 1977.

5.   Semrau, K. T., C. L. Wltham, and W. W.  KerHn,  "Energy  Utilization by
     Wet Scrubbers," EPA-600/2-77-234, November 1977.

6.   Semrau, K. T., and C.  L.  Wltham, "Wet Scrubber  Liquid Utilization,"
     EPA-650/2-74-108, October 1974.

7.   Strauss, W., Industrial Gas Cleaning, 2nd Edition,  Pergamon  Press,
     Inc., New York, N.Y.,  1975.

8.   Hesketh, H., "Fine Particle Collection  Efficiency  Related  to Pressure
     Drop, Scrubbant and Particle Properties,  and  Contact  Mechanism,"  J.
     A1r Pollution Control  Association, Volume 24, No.  10, October 1974,
     pp. 939-942.

9.   Ranade, M. B., and Kashdan, E. R., "Design Guidelines for  an Optimum
     Scrubber System," Second  Symposium on the Transfer  and  Utilization of
     Partlculate  Control  Technology,  U.S. Environmental  Protection Agency
     Publication  EPA-600/9-80-039a, September  1980,  pp.  538-560.
                                    4-14

-------
                           5.0  MECHANICAL COLLECTORS
      Mechanical collectors  comprise  a  broad  class  of particulate  control
 devices that  utilize  the mechanisms  of  gravity settling,  inertia!,  and dry
 impaction.  Performance capability is limited to the relatively large particle
 size ranges.   Because of this  limitation  and  increasing  regulatory require-
 ments, use  of  mechanical  collectors has gradually  declined and  they  are now
 used primarily  as  precleaners;  although in  certain applications,  mechanical
 collectors are  adequate.  They  are reasonably  tolerant  of high dust loadings
 and are  not  susceptible  to  frequent malfunction  if properly  designed  and
 operated.
      The principal diagnositc tool  for evaluating  mechanical  collector  per-
 formance is the static pressure drop.  Other  performance  indicators  include
 opacity and solids discharge rate.
 5.1  COMPONENTS AND OPERATION
      In a  cyclone3,   a  vortex  is  created  within  the cylindical  section  by
 either injecting  the  gas stream  tangentially  or  by  passing  the  gas  stream
 through a  set  of spin  vanes.   Due to inertia, the particles  migrate  across
 the vortex  gas  streamlines  and  concentrate near the cyclone walls.  Near the
 bottom of the  cyclone cylinder the  gas  stream makes an 180  degree turn and
 the particles  are discharged  either downward  or  tangentially  into  hoppers
 below.  The treated gas passes upward and out of the cyclone.
 5.1.1  Simple cyclones
      The simple  cyclone  consists  of  an  inlet, cylindrical  section,  conical
 section, gas outlet tube, and dust  outlet  tube.   A typical  tangential  inlet,
 axial outlet simple cyclone is shown  in Figure 5-1.
aOnly cyclones and multi-cyclones  are addressed in this manual.   For informa-
 tion on the other various types of mechanical collectors, refer to the Parti-
 culate Control Techniques Document, Second Edition1.
                                      5-1

-------
                   GAS
                   IN"
GAS OUTLET TUBE
                                         DUST OUTLET TUBE

                                    OUSTOUT
                     FIGURE 5-1.  Typical Simple Cyclone
     Particle separation is a function of the gas flow throughout the  cyclone
cylindrical diameter.   At  higher  gas flow  rates  and  smaller cylinder  dia-
meters, the  particle  inertia  is   greater  resulting  in  higher  collection
efficiency.  There is an upper  limit, however, where  the  increased turbulence
caused by higher gas velocities can disrupt  particle  concentration.
     Medium efficiency  single  cyclones  are usually less than  12 feet  in  dia-
meter and  operate  at static  pressure drops of  2  to 6  inches W.G.   Overall
collection efficiency is a function of the  inlet particle  size distribution.
                                     5-2

-------
5.1.2  Multiple cyclones
     A multiple cyclone consists  of  numerous  small-diameter cyclones operat-
ing in parallel.   The high-efficiency  advantage  of small-diameter  tubes  is
obtained without  sacrificing  the ability  to  treat large  effluent  volumes.
     The individual  cyclones,  with  diameters  ranging  from  3  to 12  inches,
operate at  pressure  drops  from  2.0 to  6.0  inches  W.G.   The  inlet to  the
collection tubes  is  axial,  and  a  common Inlet  and outlet manifold is used to
direct the  gas  flow  to  a  number of  parallel tubes.  The  number  of  tubes  per
collector may range  from  9  to 200 and  is  limited  only  by  space  available  and
by the ability  to  provide equal  distribution of the  gas  stream  to each  tube.
Properly designed  units  can be  constructed and  operated with  a  collection
efficiency of 90  percent  for particles  in the 5 to 10  ymA range.1  Factors
which may contribute to reduced performance include erosion, plugging, corro-
sion and hopper recirculation.
     Hopper recirculation in  reverse flow type  system occurs when  tubes  at
the rear  of the bank have  a slightly  lower  static pressure drop.   This  can
occur whenever  flow  distribution is  nonideal  or when the outlet tubes  are
shorter than those in the front  (a common  design).   In these  cases  a portion
of the gas  in  the  front  tubes  can  pass  out  the  bottom and then  enter  the
discharge of tubes in the back rows.
     Hopper recirculation can lead to  substantially reduced collection  effi-
ciency.  Some of  the ways  to  minimize this  problem  include segregation  of
the hopper and equalizing the lengths of all  discharge  tubes.
     Ambient air  inleakage  into  the  hopper area  can interfere with  the  dust
discharge in  the  tubes and lead to  increased emissions.   These leaks  may
occur at  hopper  flanges,  access hatches  or  through  the  solids  discharge
valves.  Severe leakage on  multicyclones  serving  combustion  systems can  be
identified by measurement  of the flue gas oxygen  content  before  and  after  the
collection.
     Plugging of  the gas  inlet  vanes  leads to partial  or total  failure  of
individual tubes  to  establish  an adequate vortex.  Partial plugging  allows
dust penetration through the affected tube. Complete plugging results in an

                                     5-3

-------
 increased  flow  of gas through the remaining tubes and an increase in  overall
 collector  static  pressure drop.   Partial plugging  of  a significant  number  of
 tubes  can  cause variation in  pressure drop from tube to tube, cause  gas  short
 circuiting through the dust  hopper, and cause an increase in  dust reentrai nment
 at  the tube  dust  outlet.
     Plugging of  outlet  tubes or the sol Ids discharge is common.  The latter
 is  generally due  to  poor  hopper discharge  practices  which  allow  solids  to
 periodically accumulate  to  the bottom of the  tubes.   Further information  on
 mechanical collectors can be  found in references 1-6.
 5.2 INSPECTION PROCEDURES
     Static  pressure  taps should be installed on  the  inlet and outlet  ducts
 so  that the static pressure  drop can be measured on each  inspection  using
 portable gauges.   Since these  taps  often  plug,  the  tap  should be  cleared
 before attempting any  measurements.   Shifts from  a baseline  static pressure
 level  may  indicate  internal   problems   and  reduced  part icul ate  collection
 efficiency.  Comparison  with  baseline levels 1s done  by  correcting gas flow
 rate and  gas temperature  data  back to levels  during the  baseline  period.
 This is done using Equation 5-1.
                              «p)      (d,nep)               Eq. 5-1
                         I ^baseline)
      where
                Q » flowrate, ACFM
                d » density factor, dimension less
The gas  densities  can be obtained  by  use of a  psychrometric  chart shown in
Figure 5-2.  To get a  density  factor, the  humidity must be roughly estimated.
     For units serving combustion  sources,  flue  gas oxygen content should be
measured on a routine basis at  the  tap  holes.  This  value  should be compared
with typical  values  for the  specific  unit  to  determine if there  is deter-
ioration of the unit.  An  increase of  more than 0.5% 02  across the collector
is a sign of excessive air inleakage.
     The rate of solids  discharge  should  be checked frequently.  Bridging of
solids in the hopper,  failure  of the solids discharge valve,  or  severe internal
problems may all  be identified by this simple technique.   All of these demand
immediate attention.
                                     5-4

-------
                                                                             CL30
               200
             300
DAY BULB TEMPERATURE IN DEGSEES F
400
400
                                                                             QOO
                FIGURE 5-2.  Psychrotnetric Chart for Humid Air
                  (Reprinted with permission of American Air
                         Filter, Louisville, Kentucky)
     Internal inspections should  be conducted by plant personnel, at a mini-
mum on  an  annual basis.   These  should  be conducted  by  personnel  qualified
in proper confined  area  entry procedures and during a major outage after the
unit has been purged  out.   The internal  inspection should  identify plugging,
erosion, and corrosion  problems.   In  some  cases it may  be advisible to use
tracers or smoke to identify gasket and weld  leaks.
                                     5-5

-------
5.3  REFERENCES

1.   Theodore, L. and Buonicore, A.J., Industrial  Air Pollution Control
     Equipment. CRC Press, Cleveland, Ohio,  1976,  p.  bb.

2.   Licht, W.. Air Pollution Control Engineering, Marcel  Dekker,  Inc.,  1980.

3.   Baxter, W., In:  Air Pollution,  Vol  II.  3rd Edition,  A.  Stern,  editor,
     Academic Press, New York 1977.

4.   Strauss, W., Industrial  Gas Cleaning. 2nd  ed., Pergamon  Press,  1975,
     p. 213.

5.   Crawford, M., Air Pollution Control  Theory, McGraw-Hill, Inc.,  1976,
     p. 236.

6.   Schneider, A.C., "Mechanical  Collectors,"  In:   Handbook for  the  Operation
     and Maintenance of Air Pollution Control Equipment, Cross,  F. L.  and
     Hesketh, H. E., editors, Technomic Publishing Co., Westport, Ct.,  1975,
     pp. 41-68.
                                    5-6

-------
                      6.0  CARBON BED ADSORPTION SYSTEMS
     Carbon adsorption is a technique for the removal of organic vapors from
a gas  stream.   The adsorption  process  is  made  possible through the  use  of
specially "activated"  carbon  adsorbent  which is  highly  porous  and  possesses
fine capillaries.  These  attributes  result in a  very  high  surface-to-volume
ratio.  Through  capillary  action  and  secondary  bonding,  the activated  carbon
can adsorb onto  its surface large quantities of volatile organics.
6.1  COMPONENTS AND OPERATION
     In a typical carbon adsorption unit, as shown in Figure 6-1, the organic
vapor  laden gas stream passes through a filter to remove particulate matter and
then through a  cooler to  reduce the  temperature to  less  than  38°C.   A  blower
then forces the gas stream through one of the carbon beds.  The organic  vapors
are adsorbed by  the  carbon  and  the purified air flows to the atmosphere.   In
order  to have continuous operation, a second carbon bed is being regenerated.
Regeneration is  necessitated when the carbon becomes  saturated with organics.
     Adsorption of a  vapor by activated carbon occurs in two stages.   Initial-
ly- adsorption  occurs at an  efficiency  approaching 100 percent, but  as  the
retentive capacity of the carbon  is  reached, traces  of the vapor  begin  to
appear in the  exit air  stream.  This  is   called the  breakthrough   point  of
carbon beyond which the efficiency of removal decreases rapidly.
     Eventually, the vapor concentrations at the inlet and exit become equal
as the carbon  becomes saturated having adsorbed the maximum  amount  of vapor
that it  can  adsorb  at the  specific  temperature and  pressure.  Figure  6-2
graphically illustrates the deterioration  of performance  as  the active sites
on the carbon surface are depleted.
     In the regeneration process, the organics are desorbed from the carbon
by passing either  steam or hot  gases  through the bed.   The  temperature  of

                                     6-1

-------
Steam
                                       Vessel  A
                        •tx-
                                       Vessel  B
                        •X-
•• •.»••».•  .•••.•••. ••.-'   •.«••••*
;•;.'.'.'>:' Carbon 'Bed;'••.•;::':•
-'.•'.'•'.''•• ••/.. '.•-.",'•*•'' ••".."•.""»'.' °
* » » • «^   * •* • • • • i , _• .* •_& * *
                                                     I
                                  Solvent
                                   Laden
                                    Air
               Cooling
                Water
                                  To Atmosphere
                                                                    Condenser
                                                                            Recovered
                                                                             Solvent
                     FIGURE 6-1.   Typical Carbon Bed Adsorber

                                        6-2

-------
    -
         Note:  This curve is an example.
               Actual time will vary for
               each application.
GOO




500




400




300




200




100

 50
      1  2  3 4  if7
             25          30          35

                 On-Stream  Time  (Minutes)
40
                                                                             45
    FIGURE 6-2.
Discharge Concentration vs. Time for a Non-Cooled Carbon-Bed
               Adsorption System
                                       6-3

-------
This desorbing medium must be  such that the carbon is heated to a temperature
higher than that at which  the organic  vapors  were adsorbed.  Saturated steam
at  low pressures  is  the most common desorbing  medium.   The steam is usually
passed through the  carbon bed in a  direction  opposite to  the  flow  of gases
during adsorption.
     The steam along with the  desorbed solvents are passed to a condenser and
then are decanted  to  separate the water and the  recovered solvent.   The re-
generation process leaves the carbon hot and saturated with water.  Sometimes
 (and in newer systems), before the carbon bed is placed back on stream, cool-
ing of the carbon is  performed.   Atmospheric air blown through the bed is most
commonly used to remove the heat from the carbon.
     Regeneration/adsorption  cycles are controlled either manually or auto-
matically.  With  automatic control,  cycles  are  either executed  for  a  set
time or vary depending on the  continuous measurement of effluent concentration
 (called "breakthrough control").  Breakthrough control utilizes monitoring of
the effluent concentration to identify  the  point where a bed is just approach-
'ing breakthrough, at which time  the  vapor  laden inlet  gas  stream is  switched
to  a regenerated bed.
6.2 INSPECTION PROCEDURES
     Excessive emissions from carbon bed adsorption systems usually result
from the typical failure  modes  of the  carbon  beds and their ancillary equip-
ment.  These failure modes are as follows:
     o  Steam valves leak;
     o  Air discharge valves  leak;
     o  Insufficient steam flow during regeneration;
     o  Insufficient time for steamout;
     o  Poor steam distribution;
     o  Condensation system failure;
     o  Collapse of carbon bed;
     o  Loss of carbon activity;
     o  Bed blockage  or channeling  due to  buildup of  lint,  dust,  polymers,
        or other materials;
     o  Uneven bed distribution; and
     o  Improper setting of system controls.
                               6-4

-------
Unfortunately, it is impossible to detect -many of these failure modes by phy-
sical inspection  and  many  of  the  inspection procedures  involved  are unduly
hazardous for the  inspector.   This  section presents  several  inspection tech-
niques that  are  presently  available  and  which  can   be  readily utilized  by
maintenance and  regulatory  agency  personnel  in determining the effectiveness
of operating carbon adsorption systems.
     Discharge gas monitoring can demonstrate that breakthrough does, or does
not occur.1  Although  not  an  exact test,  it  1s  probable that  a  system which
goes through  several  cycles of  full-load operation   without  reaching break-
through, is operating  at  adequate  adsorption efficiencies.  Determination  of
the presence of  absence  of  breakthrough requires that solvent  concentrations
in the 0-500 ppm range be continuously  measured.  Several types of instruments
are presently available  for measuring  solvents  at these  low  levels.   Among
these are:
     o  Flame ionization systems;
     o  Ultra-violet ionization systems; and
     o  Diffusion sensors.
These systems each have some advantages and disadvantages.   They  require cali-
bration for the solvent being measured  if  exact measurements are required, but
can also be used without calibration to indicate relative concentration levels.
If the output of a suitable detector is connected to  a recorder,  the adsorber
discharge concentration can be monitored  over several cycles  to  determine  if
a breakthrough occurs.   For a  facility operating at  normal production rates,
the absence of breakthrough may be taken as evidence  that the  adsorption sys-
tem is operating satisfactorily.  For successful testing an inspector must be-
come familiar with the  use of the instrumentation, safety considerations in its
use, and its application to carbon adsorption systems.
     An overall  system material balance,  taken over a  reasonable time period,
is one of the best techniques for determining overall  system efficiency.1  To
utilize a material  balance the plant must maintain accurate  records of solvent-
containing materials entering the process, and of the recovered solvent.  Re-
covered soJvent   returned to the process  is  considered as  part  of the solvent
entering the system.  The solvent concentration of all materials  entering the
process must be  known.  A  form for a  material  balance is presented in Figure
6-3.
                                     6-5

-------
              CARBON BED ADSORPTION SYSTEM MATERIAL BALANCE
Inventory Period:  From	.to
INPUT

(a)  Solvent containing materials:
     i of materials x (# solvent/1 material) * I solvent.
     1.
     2.
     3.
     4.
     5.
     6.   Total solvent in raw materials (in #)
          (-1+2+3+4+5)

(b)  Purchased solvent
     7.
     8.
     9.
    10.
    11.   Total purchased solvent (in #)
          (-7 + 8 + 9 + 10)

RECOVERED SOLVENT

    12.   Total solvent from recovery system
    13.   Solvent sold
    14.   Recycled solvent (-12-13)

% OVERALL EFFICIENCY
     100 x	(12)
           (6) + (11) + (14)
Note:  If the solvent-recovery system returns more than one solvent stream,
       a separate sheet should be prepared for each solvent stream.  Over-
       all efficiency is the average of the efficiencies for each  stream.
         FIGURE 6-3.  Overall Material Balance For Facilities Using
                             Solvent Recovery Systems
                                   6-6

-------
     Most of the information can be assembled from shipping data, and from com-
ponent data available from raw-material suppliers.  The data needed to complete
the material balance is the quantity of recovered solvent.   This data is only
available directly from the source  through  use of  totalizing flow meters which
measure solvent from the  recovery  systems.   Since not all  installations have
these flow meters  installed,  the material  balance  procedure will not be appli-
cable to every carbon bed installation.
     A third efficiency evaluation procedure can be used for systems employing
instruments (such as flowmeters) providing data  which can be used to calculate
steam to recovered  solvent ratios.^   If one  assumes a  constant  solvent input
into the system and constant  length adsorption  cycles, then if  the  ratio  of
solvent to  steam  decreases,  it is  indicative  of a  decrease  in  collection
efficiency.  This technique is  readily used  on  systems recovering  immiscible
solvents and having flowmeters on both the  solvent and water streams following
the decanter.    While  it  will  alert  maintenance/inspection personnel  to  a
malfunction of  some type in  the carbon bed  system it is,  unfortunately,  of
little aid in diagnosing the problem.
     Several types of visual inspections can also be used to help in evaluat-
ing carbon adsorption system.3  The first  visual check  that  can  be  made is  to
see if the  proper  bed  is on the adsorption  or  desorption cycle  according  to
the mode selection of the control panel.
     If the system operates on timed cycles,  then a check should be made to de-
termine if the bed actually changes at the predetermined time (the cycle dura-
tion).  At this time, the agency inspector may also inquire as to the rationale
for the  cycle  duration.  Sometimes  operating  conditions  no longer  coincide
with design specifications and cycle times have  not been adjusted accordingly
to insure adequate system performance.
     If the system utilizes breakthrough  control,  then  the  concentration sett-
ing of the  organic  analyzer  being  used  to monitor  the  exhaust  should  be
checked to  see  that it is within  reasonable limits.   If possible,  the zero
setting of the  device  (as  described in the operating  manual) should  also  be
checked.
                                     6-7

-------
      During regeneration, the  steam  pressure according to the  gauge(s)  should

 be compared to  the  design value to aid  in determining if beds  are  being fully

 regenerated.  Lastly,  a quick  visual inspection should be made for corrosion,

 especially  1n the piping  between the  carbon  beds and the  condensor and  the

 condensor system itself.a


 6.3  REFERENCES

 1.   Mlchaelis, Theodore B., "Techniques to Detect Failure in  Carbon Adsorp-
      tion Systems,"  report  by Engineering-Science,  EPA  Contract  No.   68-01-
      4146,  July 1980.

 2.   Communication  between Mr. Ted Michaelis, Engineering-Science  and Ms.
      Robin  Segal 1,  Engineering-Science, 19 February, 1982.

 3.   Communication  between Mr. Macon Shepard, Engineering-Science  and Ms.
      Robin  Segall,  Engineering-Science, 19 February, 1982.
dSome organics form acids when regenerated with steam.


                                     6-8

-------
                                 7.0  FLARES

     Flares are used  for the control  of gaseous  combustible  emissions from
stationary sources.  They  are most often employed to  dispose  of large quan-
tities of gases during plant  upsets.  In these cases flows are usually  inter-
mittent with  flow  rates  of  several  million  cubic  feet  per  hour  during  a
major upset.  Flares  have  also found use as  a control  device  for continuous
gas flows of  the  order of a  few hundred cubic  feet  an hour.   This  is a rel-
atively inexpensive  method to  control  gaseous emissions,  but unfortunately
has the  potential  disadvantage  of creating  other emissions  from  the flare
itself.
7.1  COMPONENTS AND OPERATION
     There are three  types of  flares  commonly in use  today.   These are the
elevated flare, the  ground flare,  and the  forced  draft flare.   The  choice  of
flare type  depends on  the specific  characteristics of the  gas flow  to  be
disposed of and the primary aim in flaring the gas.  Elevated flares are used
primarily in  conjunction with vapor relief  collection systems in large-scale
chemical manufacturing  or  petroleum  refining operations;  expedient  vapor
disposal rather than pollution control is the design emphasis.
     Ground flares,  which  include  conventional   flare  burners  discharging
horizontally  and  the  "low-level"  flares,   are useful  for the  disposal  and
control of routine discharges.  The low level flares  are generally less noisy,
and emit much less thermal radiation and glare.  They are also more efficient
than elevated flares in the combustion  of  gaseous wastes and thus have lower
air emissions.  Ground flares  used  alone are not suitable for large gas flows,
however, and  so  often,  low  level   flares   are  used  in  conjunction  with  an
elevated'Tlare when the possibility  of  the  sudden venting  of substantial gas
quantities exists.

                                     7-1

-------
      Forced draft  flares use  a  blower of. some type  to supply  the  air  and
 turbulence necessary  for the efficient and smokeless burning of vented gases.
 Unlike  other efficient flares, they do not require the use of steam, water, or
 an  auxiliary gas for  smokeless flaring.  Forced draft flares are  advantageous
 1n  their low operating  cost  and  reliability,  but are relatively  expensive to
 Install and can only handle a relatively narrow range of waste gas  flow rates.
 7.1.1  Elevated Flares
      The modem  elevated  flare   system  1s  made  up  of  several  components
 Including the  flare tip,  some type of  gas  trap  directly below the  tip, a
 pilot and  Ignition  system  at the  top  of  the  flare tip, and the stack and  its
 support.  When  smokeless burning  1s required, a  steam  Injection  system must
 also  be provided at the top  of the  flare.  Water seals to prevent the spread
 of  fire and knockout  drums  are  also  usually required  for  safety  reasons-
 Figure  7-1 shows a  schematic of a typical elevated flare system.
      The flare  tip  where  waste   gas  combustion occurs  must  have  excellent
 flame holding ability and mixing characteristics to enable 1t to  operate over
 a wide  range  of turndown ratios.   This 1s ensured using multiple continuous
 Pilots  around the combustion tip  and by providing  a  flame stabilization ring
 on  the  tip (see Figure 7-2).
      Smokeless burning  for elevated flares  1s  achieved using  special  flare
 tips  which inject water, natural gas or steam  into the flame thereby increasing
 air-gas mixing  and  ensuring complete  combustion.   Water  injection  has  many
 disadvantages and natural gas injection 1s cost effective only 1n cases where
 the gas has  no value,  so  steam  is most  commonly used  to achieve  smokeless
 burning.
      There are two  basic steam Injection  techniques used 1n elevated flares.
 In  one  method  steam is Injected from nozzles on an external ring around  the
 top of  the tip  (Figure  7-2).  In  the  second method the  steam  is  Injected by
 a single nozzle located concentrically within the burner tip.
      In recent years  regulations  have  required flares  to be  smokeless   for
large turndown  ratios.   To ensure  satisfactory  operation under  varied flow
conditions, these two types  of steam  Injection  have been  combined  into   one
tip.  The  internal  nozzle  provides steam  at low gas  flow rates  and the

                                     7-2

-------
L,
Cy
(2]
                                      Flare Burner and Location
                                      of Fluid!c Seal
                                      Gas Trap
                                      Riser Sections
                                      Entry, Disentrainment
                                      or Water Seal
                                      Ladders and Platforms
 FIGURE 7-1.  Integrated Flare Stack  Components1
                     7-3

-------
Utility Field Flare Tip

                    Flame Retention Ring
                        •Pilot Assembly
                                     Endothermic Field Flare Tip
       Smokeless Field Flare Tip
                                                             End other mi c
                                                             Assist Gas
                                                             Supply
                                   Steam Distribution Ring
           FIGURE  7-2.  Flare Tips from John Zink  Company!
                                7-4

-------
external jets are available at high flow rates.   Several  other special purpose
tips are currently  in  use including ones that inject  steam  and air and ones
that inject gas to  raise the heat value of the waste gas stream.
     The rate of  steam injection to the  flare tip  is  controlled manually or
automatically.  While  automatic  control   is  usually  not  mandatory,  it  is
preferred because  it  reduces  steam  usage,   greatly  reduces  the amount  of
Smoking and minimizes  noise.   Automatic systems use flow measurement devices
with' ratio control  on  steam.   Since the flow rate measurement cannot include
the variables of degree of saturation  and  molecular weight, the  ratio controls
are usually  set  for  some  average  hydrocarbon   composition.   It  is usually
necessary to have a  fixed  quantity  of steam  flowing at all  times to cool the
distribution nozzles at the tip.
     To prevent air migration into the flare stack as  a result of wind effects
or density difference  between air and  flare  gas, a  continuous purge gas flow
through the flare system 1s maintained.  The  system can be purged with natural
gas, processed gas, inert gas,  or nitrogen.   To  reduce  the amount of purge gas
required and to keep air out of  a flare system, gas trap devices are also used
and are normally located in the  stack  directly under the flare tip.  One type
of gas trap commercially  available  is the molecular seal (Figure 7-3).   This
type trap may not  prevent air from getting  in  the stack as a  result  of gas
cooling in the flare headers.  Instrumentation systems are available to auto-
matically increase the purge rate to prevent air  from entering the stack dur-
ing rapid gas  cooling.  Another type  of gas trap is  the Fluidic  Seal  which
utilizes baffles to create a velocity gradient  between  the  waste gas and the
ambient air to prevent air migration  into the flare  stack.   This seal  weighs
much less than a molecular  seal and can be placed much closer to the flare tip,
     The pilot burners ignite the outflowing gases and keep them lit.  Usually
three or four  pilot burners are used.  A  separate  system  must  be provided
for pilot burner  Ignition  to  prevent   flare failure.   The usual  method  is to
ignite a gas/air mixture 1n an Ignition  chamber  by  a  spark.   The flame front
travels through an  Igniter tube  to  the pilot  burner at the  top  of the flare.
This system permits  the igniter  to be  set  up   at  a  safe distance  from the
flare, up to  100  feet,  and  still  Ignite  the  pilots  satisfactorily.   The
whole device is  mounted  on an  ignition  panel   and  set  up   in  an  accessible
spot on the  ground.  The Ignition panel must be  explosion proof, have an

                                     7-5

-------
unlimited  life, and be Insensitive to all weather conditions.  On elevated
flares, the pilot flame 1s usually not visible and an alarm system to Indi-
cate pilot flame failure is desirable.   This is  usually done by a thermo-
couple in  the  pilot burner flame.  In the event of flame failure, the tem-
perature drops and an alarm sounds.
 Outlet to Flare Burner
  Outlet to Flare Burner
                      Clean-Out
  Inlet from Flare Riser
National Air Oil NDS Double Seal
(Patent applied for)
 Inlet from.Flare Riser
John Zink Molecular Seal
(U.S. 3,055,417)
                     FIGURE 7-3.  Air Reentry Seals1
                                  7-6

-------
     Water seals  and  flame arrestors  are  used  to prevent a  flame  front
from entering the  flare system.   Knockout  drums  are located at  or  near
the base  of elevated  flares  to  separate  liquids  from  the  gases  being
burned.
7.1.2  Ground Flares
     A ground flare consists of  a  burner  and auxiliaries, such as, a seal,
pilot burner and igniter.  Two types are found; one type consists of con-
ventional burners  discharging  horizontally  with  no enclosures.   These
flares must  be  installed  in  a  large  open  area  for  safe  operation  and
fire protection.   If  the ignition  system  fails they are not as capable
in dispersing the  gases as  an  elevated flare.   For these  reasons  this
type of ground flare has found only limited applications.
     Ground  flares may  also consist of  multiple  burners  enclosed within
a refractory shell  as  in the  recently  developed "low level"  flares  (Fig-
ure 7-4).  The essential purpose of a low level  flare  is complete conceal -
"lent of the flare flame as  well  as smokeless burning at a  low noise level.
The flared gases are  connected  by a manifold  to a  series of burner  heads
which discharge the gas into  a  refractory enclosure.  Mixing of the  gas
and air 1s accomplished  by  a  series of multi-jet  nozzles.   Combustion  is
obtained with little or no  steam because of the turbulence and temperature
of the burning zone due to  the natural  draft and the enclosure.   This  type
of ground flare can be  designed  to handle most plant flare occurrences  and
then the remaining  large releases can  be diverted to an  elevated flare-
Figure 7-5 is a schematic  showing how  such  a  system  works.   This type  of
integrated flare system  is becoming quite common,  especially in  populated
areas.
7.1.3  Forced Draft Flares
     The forced draft  flare uses  air provided  by  a  blower to supply  pri-
mary air and turbulence  necessary  to provide  smokeless burning  of relief
gases without the  use  of steam.   Figure 7-6  shows two  common designs  of
forced draft flares.  This  type of flare combines  smokeless  burning  with
low operating cost and reliability because only pilot gas and electricity
are required.  The flame is also stiffer  and,  because of the forced draft,
1s less affected by the wind.  A blower flare should have  an  automatic air
turndown device to prevent excess air   from quenching the flame and creat-
ing smoke if the flare  gas  rate is reduced.   Variable  speed  blowers  or

                                   7-7

-------
baffles coupled to flow sensing devices  have  been used on these flares to
extend their turn-down ratio.   These  flares are  used  mainly to provide
smokeless burning where steam  1s not  available.

k
1
III

-p
/ 111 111 ;

a
0 0
OJfjO
ol|o o^o
o o o o
o|o ojfjo



1 '





(S)
r\ ,
N
I I




I
$



i

\


1
N
i
I
i
1 i
I 1 I1








^
^ '
^ '





ii •











@

* *
,




















a t





ii i












/
T- ;
a t

















_— • — •
91
J
\
• V
D l



•«^^^—


II











	 • 	

". ''•
i £



^MM^^^H
•- "

-*

Item Descriotion

1 Patented Jet Mix Tips
2 Flare Gas Risers
3 Flare Gas Header(s)
4 Flare Gas Connection^ )
5 Combustion Chamber
1 1 6 Refractor/ Lining and
1 1 Anchors
1 1











7 Safety Fence (Collapsible)
8 NSFP (Pilots with Igniter
Tubes
9 Sight Ports


/r\
^•"1
r~i~~^v


	 	 	 ^ 	
   FIGURE 7-4.  Ground Rare (from National  A1r01l  Burner Company)1
                                 7-8

-------
 7.2  INSPECTION PROCEDURES
      The Initial and by far the most informative procedure in  a  flare  in-
 spection is the  visible  emissions  observation.   The presence or  absence
 of visible  smoke  is  a  primary  indication of the  flare's  combustion  effi-
 ciency.  Almost all flares in  use at the  present  time will be of smokeless
 design and  therefore  should   have  essentially  no  visible  emissions.
     Pilot  Gas
     Line
                                Elevated Flare Burner
Flare Gas
Line
                                 Diversion
                                 Seals
                                  Control System
     FIGURE  7-5.   Ground  Flare  and Elevated Flare  Connected  by a Double
                  Stage  Water  Seall
                                    7-9

-------
     If smoke  1s  visible,  this  1s  usually  an  Indication  of  some  sort  of
malfunction.  An exception  occurs 1n cases where the  flare  1s  burning waste
gases from a major plant upset and the gas  flowrate has exceeded the practical
limit of  smokeless  burning  for  that particular  flare.   The  most  critical
determinant of smoke production 1s the amount and distribution of oxygen  1n
the combustion  zone.    Therefore  a  smoking  flare  might  be  caused  by:
     1.   Lack of sufficient  steam Injection  (to provide turbulence and
          oxygenated compounds) at the combustion zone,
     2.   Lack of sufficient  water Injection at the combustion zone, or
     3.   Lack of sufficient  air from blowers 1n forced draft systems.
                          Combustion
                          Air Inlet
                                 Flare Gas
                                 Inlet
                                                                    Combustion
                                                                    Air Inlet
  a.  Biaxial Forced Draft Unit
b. Coaxial Forced Draft Unit
            FIGURE 7-6.  Two Designs for Forced  Draft  Flare  Systems*
                                      7-10

-------
Comparison of the values  from steam or water pressure gauges (if available)
with manufacturer's design  values may provide  an  indication  of  whether
Sufficient pressure is available for smokeless combustion.
     Unfortunately, flares do not otherwise readily lend themselves to reg-
gulatory and routine maintenance Inspections.  Some further cursory inspec-
tion procedures include being aware of malodorous compounds in the flare's
vicinity and examining the ducts and  pipes  carrying the various gases,  air
and steam for obvious leaks and signs of deterioration.
     A potential future Inspection procedure for flares  1s the remote  sens-
Ing of the components  1n  the gas stream exiting the  flare.   This  method
may provide  a  means of determining  the  combustion efficiency of  flares
and the actual  quantities  of pollutants emitted to the  atmosphere.   For
an example of the  application  of remote sensing on flare  emissions,  the
report by W.  F.  Herget  et  al   entitled  "Development  of  Flare  Emission
Measurement Methodology"  is recommended.
7.3  REFERENCES
1.   Klett, M. G. and J.  B. Galeski,  "Flare Systems Study," report  by  Lock-
     heed Missiles  and  Space Co.,  Inc. for  U.S.  EPA Research  Triangle
     Park, NC, Publication No.  EPA-600/2-76-076, March 1976.
2.   Communication between Mr.  Glenn  Draper, Engineering-Science  and Ms.
     Robin Segal!, Engineering-Science,  18 February 1982.
                                   7-11

-------
                           8.0  AUXILIARY EQUIPMENT

     There are  three  main  types  of  equipment  auxiliary  to  air  pollution
control devices.   They are  the  (1)  ventilation  system  which  includes  the
hoods and ductwork  leading to the air pollution  control  device, (2)  the fans
or blowers associated with the control device, and (3) the continuous monitors,
if any,  installed  after  the control  device  to  monitor  the  effluent.   This
chapter discusses  the counter-flow  inspection  procedures  for  these  types  of
equipment.
8.1  VENTILATION SYSTEM
     The ventilation  system  is  simply  the  hood and ductwork  leading  to  the
air pollution  control  device.   Since there  are  no moving parts it  is  often
erroneously assumed  that  nothing  can  go  wrong.  Actually  the ductwork  is
prone to  serious  air  inleakage  due to  erosion  and  corrosion.  Also,  under
certain circumstances  solids  can  build up,  thereby  restricting gas  flow.
     Evaluation of the  ventilation  system is done by  observing the  physical
condition and by measuring static pressure and gas temperature profiles.  For
combustion sources  an oxygen  profile  is  very  useful.   These  profiles  are
acquired by measuring the various  parameters  at three or more points  along
the ductwork.   The measuring points could be simple 1/4"  O.D. static  pressure
taps or 1" to  2"  ports equipped with threaded caps.   They do  not have  to  be
equally spaced.
     The measurement  results  are plotted  on a  graph  as a  function of  the
equivalent linear distance from the control  device.  As  shown in Figure 8-1,
the static pressure (on systems with the fan after the control device)  should
become progressively less negative  as one moves  away  from the control  device
toward the process.   The  slope of  the  line  will be basically  linear  if  the
equivalent lengths  of elbows  and duct  expansions  are used  (calculation  of

                                     8-1

-------
0
0
3 - 2
0)
^ - 4
o n
"1 - 6
0)
s_
S - 8
4)
^ -10
O
1 -12
»
1 1 »
^
. , 	 	
1 1 1
• i i


\ Mechanical
» Collector
Scrubber
i i

100 200 300 400 500
Equivalent Length, Feet
1



i_ 	
_
_
Fan -
"
600 700
FIGURE 8-1. Example Static Pressure Profile (System Having a
Mechanical Collector and Wet Scrubber)
15
« 13
£ 11
4->
5 9
c
 i.

-
-
4-ff
+*r—
1 1 1 1
i



i
I
i»


1
i
—
f**^

•»
i
400

200
                                                                  OJ



                                                                  (O
                                                                  s_
                                                                  0)


                                                                  
-------
equivalent lengths is discussed in Reference  1).   A  sharp change  in the static
pressure profile  slope suggests  either  partial  blockage  of  the  duct  or a
large air infiltration  point  due  to a hole,  weld failure,  or  an  open clean
out trap door.   Air  infiltration should be apparent also on the gas temperature
profile.  This  should  increase  directly proportional to  the  distance  away
from the collector  (see Figure 8-2).  A  discontinuity in  the  line3 indicates
the general  location of an air inleakage.  The flue gas oxygen content should
not change  substantially;  however,  most  systems  have  some  degree  of  air
infiltration which leads to a gradually increasing Q£ level  as one moves away
from the  process  (Figure 8-2).   A  sharp  increase  in  the  flue  gas  oxygen
content is indicative of air inleakage.
     One common cause of air inleakage is holes caused  by erosion.  Particulate
matter laden gas  streams  moving  at  high velocities  (3500  to 5000  feet  per
minute) can be  quite abrasive.  Abrasion normally occurs at elbows and anywhere
the gas stream shifts direction sharply.
     Uninsulated ducts  are  susceptible to corrosion, especially  if  the  gas
stream has a high concentration  of acid vapors.  Moisture on  the  relatively
cool inner  surface  of  the  ducts  absorbs the  acidic  compounds  and  promotes
corrosion.  As air  infiltration  increases, the overall  gas stream is cooled
by the ambient air,  thus accelerating corrosion.
     Hoods should be  inspected to see if they have been damaged  by  overhead
cranes or other moving  equipment.   It is also common to  find that operating
personnel have moved the hood to  gain  better access to the  process equipment
or have  adjusted dampers   controlling  the  gas  flow.   The  inspector  must
determine whether the  present  hood configuration provides adequate  emission
capture.  A velometer  or  small pi tot tube may be used in measuring  the  gas
velocities around small  hoods,  while  for  large  hoods visual  observation may be
the only method available  to  evaluate captive effectiveness.   In  some cases
it is possible to evaluate  hoods  using static  pressure readings  one to three
duct diameters  downstream from the hood.  This  static pressure  is  related to
the gas flow rate as shown in Equation 8-1.
Negative pressure systems only.

                                     8-3

-------
          Sph - (1-0 + Fh)Vp                                         Eq. 8-1
     where
          Sp  - hood static pressure,
          Fn  » hood entry loss factor.
          Vp  * velocity pressure.
     The hood entry loss  factor is constant  for a given  hood  configuration
(see reference 1  for  typica-1  values).  The  velocity  pressure calculated  in
this manner can be  used to  estimate gas flow.  Use of this  approach enables
Inspection/maintenance personnel to "baseline" a hood.   This method, however,
only indicates the overall  gas  flow rate; 1t  does  not  conclusively show that
the contaminants are being effectively captured.  The  latter is  achieved only
when the overall gas flow rate and hood location are proper.
8.2  FAN EVALUATION
     Inspection of  the  fan requires  only  a  modest investment  of  time  and
provides very  valuable  diagnostic  information which  can be  used  later  to
narrow the  scope  of the inspection.   The  procedure described  in  this section
is useful for evaluation of relatively small  direct- and belt drive fans with
rated gas flow  rates  of less  than  50,000  acfm.   Larger,  more  complex systems
may be  evaluated  using information available  in  standard texts  such  as  Fan
Engineering, (Buffalo Force, Co, 1980).   The Counterflow Inspection Procedure
utilizes the data listed in Table 8-1.

                          TABLE 8-1.  FAN INSPECTION
             1.   Physical condition of fan housing
                  a.   Vibration
                  b.   Corrosion
                  c.   Vibration/expansion sleeve integrity
                  d.   Damper position
                  e.   Hatch access
             2.   Fan operating conditions
                  a.   RPM
                  b.   Gas temperature
                  c.   Sheaves ratio
                  d.   Slippage squeal
                  e.   Fan motor current
                                     8-4

-------
8-2.1  Physical Condition of Fan Housing
     The first  item  to be checked is fan  vibration,  and this is done simply
by observing  the  fan.   The  purpose  of  this initial  check  is  primarily  to
indicate whether  the  inspection  can  continue.    In   certain   very  unusual
circumstances, the  vibration may  be so  severe that  fan  disintegration  is
imminent.  Such an event is  life threatening,  and accordingly the inspection
should be halted  immediately and  plant management  should be advised  of the
fan conditions.
     Most small fans have flexible  sleeves on the  inlet and  outlet  ducts.
These serve to  reduce the transmission  of  normal fan  vibration  to the duct-
work and  also  provide  for  thermal  expansion.   Cracks  and  tears in  these
sleeves are quite  common.   If  these  occur on the  inlet side,  cold  ambient
air is drawn  in reducing the quantity of gas pulled through the process hoods
and air  pollution  control  systems.   This  may  result in  increased  fugitive
emissions at  the process and increased fan energy use.  If the fan happens  to
be located before the  control device, then  both the  inlet and outlet  sleeves
are important.
     Corrosion on  the  exterior of  the housing  may  be  indicative  of  gas
temperatures  below the  acid  vapor and/or  water vapor dewpoints.   If  this
problem does  exist,  the ductwork leading to the fan  from the control  device
will also suffer  from  corrosion,  perhaps  to the extent of  significant  air
inleakage.  The latter  is  easily  identified  by  a  lower than  expected  gas
temperature coupled  with higher than normal flue  gas oxygen contents  (com-
bustion sources only).   Most oxygen  analyzers  are relatively easy to  use and
are reasonably accurate  (+_ 1 percent 02).
     On a routine  basis, plant  personnel  should inspect  the fan wheel  for
erosion, corrosion,  and  material  accumulation.   It  is  often possible  to  con-
firm that this  is being  done  by  examining the  hatch  or the fan  housing.
8.2.2  Fan Operating Conditions
     The most  important  fan operating  parameter  is the  rotational  speed
measured in revolutions  per minute (rpm).  The gas  flow rate through  a  given
system is directly proportional  to the rpm.  A rough estimate of the  rpm for
belt-driven fans can  be obtained simply by multiplying the fan motor rpm by

                                     8-5

-------
the ratio  of the  fan  and  motor  sheaves  as  indicated  in Equation  8-2a.
                 Fan rpm » (motor rpm) motor sheave diameter         Eq. 8-2
                                        fan sheave diameter

Obviously, the measurement of sheaves is subject  to  some error  since it must
be done by looking  through a  belt guard.  (Caution:  the  belt guard should not
be removed to obtain  a more accurate measurement).   Belt  slippage  can also
lead to a  rotational  speed  several  hundred rpm  lower than estimated.   This
slippage is usually detectable by an  audible  squealing and in extreme cases,
poor belt tension.
     A more accurate measurement of  rpm can be done by use of either a manual
tachometer, a phototachometer,  or a strobe  type tachometer.  Use of a manual
tachometer is very  straightforward assuming that  there is a convenient access
port to the fan  shaft.   Often  the configuration  of the  belt guard precludes
use of the manual tachometer.  In such cases a phototachometer may be of use.
This device senses  and counts  pulses  of reflected light.   The  most  accurate
readings are obtained  when a piece  of  reflective tape  (usual'y  supplied  by
the instrument manufacturer)  is  fixed on the  fan shaft.   It is  possible  to
read integer multiples of  the  rpm if there is more than  one reflective site
within the field of  view of the  sensor.   The rough estimate derived  by the
sheave ratio is  generally  useful  in  confirming that  the  lowest  reading is  in
fact the actual  rotational speed.
     Fan rpm on  simple fans cannot increase unless the source personnel either
replace the motor or the  sheaves.  Since an increase  in  rpm normally results
in a  greater  gas flowrate, the performance of certain types of collectors,
electrostatic precipitators and fabric filters may be adversely  affected.   If
the tip speed has exceeded permissible limits  (specified by the manufacturers)
fan disintegration  may result.
     For most fans  used with air pollution control  systems the total horsepower
required (termed the "brake  horsepower") increases as the  gas  flow  rate  in-
creases.  This is illustrated in Figure 8-3 which  is a typical  set  of curves
for a radial blade fan.   As  the operating  point  in  Figure 8-3  shifts to the
right it is apparent that  there  is  a  non-linear  increase in the brake horse-
power.
dThe rpm of direct-drive fans does not vary.

                                     8-6

-------
                                   7     I    S    10
                                  us riow RATE. SCFH x 10'
n
    12
           FIGURE 8-3.   Fan Curves and System Characteristic Curve

     If the brake horsepower could be accurately calculated  based  on simple
measurements of  fan  operating  conditions,  it  would be  possible to  make  a
reasonable estimate  of the gas  flow  rate through  the system.   Unfortunately,
this is usually  not  possible'since the brake horsepower  is  dependent on the
power factor which is neither easy to measure nor constant.  The relationship
between brake  horsepower  and  motor  operating  parameters  is  presented  in
Equation 8-3.
Fan'brake horsepower * current  x voltage x power factor x 0.746      Eq. 8-3
                        (amps)    (volts)
     Due to  the  direct  relationship  between motor current  and the brake
horsepower and the basic  nature of  the fan curves  (see  Figure 8-3),  it  is
possible to identify significant changes  in  gas  flow rate using only the fan
motor current which is relatively easy to measure using an inductance ammeter.
Normally, an electrical cabinet must be opened to reach  wire not covered by
conduit piping.  If  the motor is a three phase type, the current in all three
wires should  be  measured  and  the  average  current  computed.   (CAUTION:
INSPECTORS SHOULD NEVER  OPEN  ELECTRICAL  CABINETS  DUE  TO   POTENTIAL  SHOCK
HAZARDS.  THE  PLANT  ELECTRICIAN  SHOULD  MAKE  ALL  SUCH  MEASUREMENTS.)   The
motor horsepower is  normally used in fan performance evaluation.
                                     8-7

-------
     Since the  brake  horsepower  1s also a  function  of  gas  temperature, the
measured motor  current  must  be corrected back to  standard  conditions before
comparison with  the  baseline  motor current.  This  conversion  may  be  done
using Equation 8-4 with the data presented  in Table 8-2.
Motor Current (@STP)  - Motor Current (@Temp) x (Table 8-2 Value @Temp) Eq. 8-4
     If the corrected motor current has shifted more than 20% from the baseline
value, there  has  probably been  a  significant change  1n the  gas  flow  rate
through the  air pollution control  system.   The  actual  prevailing  gas  flow
rate can be measured  using EPA Reference Method 2 procedures.

                TABLE 8-2.  FAN DATA,  TEMPERATURE CORRECTION*
Temp
°F
20
40
60
80
100
120
140
160
180
200
220
240
260
280
300
Factor
0.91
0.94
0.98
1.92
1.06
1.09
1.13
1.17
1.21
1.25
1.28
1.32
1.36
1.40
1.43
Temp
°F
320
340
360
380
400
420
440
460
480
500
520
540
560
580
600
Factor
1.47
1.51
1.55
1.59
1.62
1.66
1.70
1.74
1.77
1.81
1.85
1.89
1.92
1.96
2.00
             dFrom  Basic Energy/Environment Analysis^,  NAPA
             Information series 67,  by C.  Heath,  August  1978.
8.3  REFERENCES
     1.   American Conference of Governmental  Industrial  Hygienists, Indus-
          trial Ventilation. Sixteenth Edition, 1980.
     2.   McDermott, H. J., Handbook of Ventilation for Contaminant Control.
          Ann Arbor Science, 1976.
                                      8-8

-------
                            9.0  PROCESS EQUIPMENT

     The purpose  of the  process  inspection  is  to  answer  questions  and  to
confirm conclusions  reached in  earlier  steps.   However,   if  the  source  is
subject to  special  state/local  regulations  or  to  New  Source  Performance
Standards or to  National  Emissions  Standards for  Hazardous Air Pollutants,
compliance with  recordkeeping  and  monitoring requirements   should  be  checked
first, using a DSSE-published  series of  documents  which  includes checklists
and associated information  on recordkeeping  and monitoring requirements.
     In the process  inspection the  inspector can  seek answers to questions
which arose earlier.  He  can address problems which  could  be  individually  or
collectively responsible  for nonoptimal performance.   Questions he might  ask
include but but are not limited to:
     1.   Has the production rate increased (higher mass loading c.nd/or
          gas flow rate)?
     2.   Have the raw materials  and/or fuels changed to the extent that
          effluent characteristics are different?
     3.   Has the process equipment deteriorated to the extent  that emissions
          are affected?
     4.   Have changes in operating conditions resulted in  more difficult
          collection problems  (e.g. particle size decreases)?
The process inspector  should  begin at the control centers  where  the  process
monitors are located, to  look  for signs  of changes in operating  conditions  and
to observe current operating/maintenance practices.  At these centers,  process
operating data are  available;  a process flowsheet  is generally posted on  the
control panel; plant  operators are  generally  nearby;  and  the  subdued  noise
level  is conducive to technical discussions.
     While in the  control room,  the inspector should  seek out  the  process
monitors and/or records most pertinent  to  the compliance questions.   Example
"inspection points" for six  source types  can be found  in Table  9-1;  several
generalizations can be drawn from this table.

                                     9-1

-------
     1.   In most cases,  the inspector can confirm increased production
          rate by using data available in the control  room;

     2.   The inspector can confirm raw material  changes  by  inspecting
          records kept either  in  the control room  or in the  administrative
          offices;

     3.   The inspector cannot  easily confirm process  operational  changes;

     4.   For batch operations, it 1s necessary to observe the  equipment since
          little useful   Information  is  available  in  the control  room;  and

     5.   For other processes,  it  is  possible to  Identify changes  in  operating
          conditions,  but the  significance of the changes  is hard to deter-
          mine.

Follow-up questions remaining after the inspection of  records and monitors  in
the control  room can  be  quite  time consuming due  to the  cyclic  processes and

to distances between inspection points at large plants.   The follow-up can  be
guided by the items listed in  Table  6,  but  the  inspector  is  encouraged  to

develop more extensive lists for  specific  plants.  The presence of  fugitive
emissions should be noted.
                                    9-2

-------
 TABLE 9-1.   EXAMPLE  INSPECTION POINTS, COUNTERFLOW INSPECTION  PROCEDURES
   Type of
Industry/Source
                       Inspection  la the
                   Office and/or Control Room
                                                              Inspection of
                                                            Specific Equipment
.1.  Con f Ira the Hates of  Production end/or Cenerstlon
Sulfuclc
•eld

Aaphalt
Utility or
Induttrlel
boiler

Ceaent

Refinery
              Check scld production  records, and
              observe acid flow rat* Indicator.

              N/A
               Check megsvatt  generation and steam
               production ratee.
                                                      N/A
Observe number of  betchee shipped
per hour of plant  operation.

N/A
               Check rev aaterlal  feed rate records.    N/A

               Check throughput recorde on catalytic    N/A
               cracker.
 2<  Confirm Raw Material Changes
Asphalt        N/A
Utility or     Check dilly records of eoalyiee:
Industrial     Z ««h,  IS, Btu content, ash fusion
boiler         tenp.

Sulfurlc       Check records of feed content!  high
acid           levels contribute acid mist.
                                                      Perform lab  teate to determine coal
                                                      gradation and surface moisture
                                                      percentege.

                                                      Tske semple  for  later analyses.
                                                      K/A
 Refinery
                Check  production Inventory recorde.

-------
vo
                           TABLE 9-1.   EXAMPLE  INSPECTION POINTS.  COUNTERFLOW INSPECTION PROCEDURES  (cont'd.)
                                 Typ« of              Inepection la th«
                              Industry/Source     Of(let and/or Control Room
                                              Inspection of
                                            Specific Equipment
                              3.  Confirm Process Operation*!  Changes
                              Sulfurlc
                              •eld
                              Secondary
                              bras* and
                              bronie
                              Utility or
                              Industrial
                              boiler
Check cacalyet bed leap and air  flow
rate to catalyst bed If SOj Monitor
valuta are  high.  Check acid concen-
tration teap  end flow to absorber*.

Check record! for percentage of  line
in alloy and  for pouring teap.
 Check air preheater  exit
 temperatures.
H/A
Check for grease and oil on scrap tod
for operational practices such as
Maintenance Integrity of slag load.
Check hood caption velocity in
furnace aree.  Determine If line 1*
added before furnace teap le maximum.

H/A
                              4.  Confirm Process  Equipment Deterioration
                               Sulfurie
                               acid
                               Secondary
                               braa*  and
                               bronie

                               Utility or
                               Industrial
                               boiler

                               Refinery
 Observe SOj concentration monitor*!
 check  for inactive or poisoned
 catalyst if SO] is high.
 N/A



 Check excess air level by meana of
 Oj  readings.


 N/A
 Check pressurs drop across  mist
 eliminator.  If low, check  for
 shortclrcuitlnii if high, check  for
 plugging.

 Check hoode and ductwork'for
 physical damage end caption velocity!
 Check to ae*  if stoker boiler
 draft above fuel beds has >0.10"
 negative pressure.

 Observe flow rate.

-------
                      10.0  VISIBLE EMISSION OBSERVATION

     Plume opacity 1s evaluated in order to:  (1) confirm compliance with
applicable regulations, and  (2)  to diagnose performance  of air pollution
control equipment.  This  chapter does not  deal   directly  with  compliance
determination, rather it emphasizes the diagnostic aspects of the visible
emission observation.  For those  requiring  detailed  information  on legal
requirements pertaining to opacity observations,  it can be found in refer-
ences 1 and 2.
     Plume opacity is the  degree to which  ambient light  is  absorbed  and
scattered while passing through the  plume to the observer.  The usefulness
of the plume opacity  concept  is  due to  the  fact  that  the extent  of light
attenuation due  to  absorption and  scattering  is often  directly  related
to the particulate mass concentration  in the plume gas stream;  an  increase
in plume opacity normally indicates an increase  in the effluent mass con-
centration.  Opacity  is an especially useful, indicator for wet scrubbers
and electrostatic precipitators  since  the  particles  which are collected
the least efficiently are  in  the  particle size range which attenuate light
most effectively.  As the  performance of these control devices begins  to
degrade, the plume  opacity  increases  significantly.   Thus,  opacity  can
be early symptom of developing particulate collection  problems.
10.1  OBSERVATION PROCEDURES
     Incorrect visible emission observation  procedures can result  in sub-
stantial  errors.  Accordingly, it  is important to  adhere to the  procedures
incorporated into the U.S. EPA Reference Method 9 (and most other  similar
regulations).  The basic requirements are as follows:
     1.  Observations must be made by a  qualified observer who
         has had formal  training  in proper procedures.
     2.  Readings should be made  in the  densest portion of the
         plume which is free of condensed water.

                                     10-1

-------
     3.  Readings must be made with the son at the observer's back,
         within a quadrant of 140°.
     4.  To the extent possible, the observer should be approxi-
         mately 90° (geometrically normal) to the plume travel.
     The opacity observations should  be made  every  15  seconds for a total of
at least six  minutes.  Longer time periods are  often  desirable  in  order to
ensure representative operating conditions for the  control  device.  If emis-
sion spikes or "puffs" are observed on an  internrittant  basis, the observation
period should be long enough to characterize  the frequency.
     A complete description  of  the observation  conditions should  be  made at
the time of the observation  (not written  from memory at a later time).  This
should include the position of the observer, the type  of  background,  weather
conditions, and  plume  characteristics.    One convenient  form  for  opacity
observations, developed by Eastern Technical Associates under contract to the
U.S. EPA, is included in Appendix C.
     The position of the observer should be characterized in sufficient detail
to estimate the angle  of observation.  On sources  with high  stacks  or with
rectangular vents (such as roof monitors), it is  possible to get an apparent
opacity which is above  the actual  opacity due to the  angle  of  observation.
This is possible since the light coming to the observer  is  passing through  a
wider section of the plume as shown in Figure  10-1.
                                 1200*
             1200*
       FIGURE  10-1.    Extended   viewing   pathlength   on   an  elevated  source
              (Path  I   is  actual  diameter,  Path  L1   is  apparent  diameter.)
                                     10-2

-------
     The difference between the apparent  opacity  and  the  actual  opacity for a
30 foot diameter stack is shown in Figure 10-2 for the observation angles of
30°, 45°, and 60°.   It  is  obvious that the  angle  of observation can have a
significant effect,  especially  if  the  opacity readings  are  made at various
locations.  In  such  a case the  change in opacity levels  from the baseline
level may be  due  at  least  partially  to a change in  observer location.   For
this reason, diagnostic  opacity  readings  should  always  be corrected back to
an observation angle  of 0.
               10    20    30    40    50    60    70    80    90    100
                                Real Opacity, %

FIGURE 10-2.   Effect  of  the Angle  of  Observation  in the  Observed Opacity
                       for  a Stack 30  feet  in  Diameter
                                     10-3

-------
     The position  of  the  source relative to  other possible sources  of  parti-
culate and/or steam should be recorded on .the observation form.  There  should
be  no question that the observed  opacity is  for the source of interest  only.
Conditions which may  have influenced the readings such as fugitive  road  dust
and intermittant steam releases along the line of sight should  be  noted  on the
observation form.   Obviously,  these values  should  be  ignored when  analyzing
the data to determine if opacity levels have shifted from the baseline levels.
     Plume characteristics must be  carefully documented.   Certain  effluent
gas streams  contain  volatile  compounds   or  reactive  compounds.   During the
rapid cooling which  occurs as  the effluent leaves the  stack,  condensation of
the materials  can  lead to  substantial   quantities  of  aerosol which did not
exist as particles  in the control  device.  Plumes composed primarily of  such
aerosols are termed "secondary plumes" or "detached plumes".  The presence of
a secondary  plume  usually  does not imply control  device  operating  problems.
Secondary plumes often  have  very  low  opacity at the  stack  exit  (between 0%
and 10%) with  very high opacities 1 to 5 stack diameters  downwind.   For  this
reason the  plume appears  to  be "detached" from  the stack.   Secondary  plumes
do  not revaporize  like  steam plumes.
     The presence  of condensed  water in the  plume  usually  can  be  easily
recognized by the  experienced  observer.   If  such a "steam plume" exists, the
readings should be made either before the water condenses or after  it evapor-
ates to leave a residual plume; the location chosen by the observer  should be
noted on the form.   Reading in the region after evaporation is often the  only
option available; however, it has no diagnostic  value since the observed path-
length is never the  same  as  that  used  in the baseline  opacity observations.
     Occasionally,  there js.  some  question as to  whether  the  steam plume has
actually dissipated and the  observed plume  is the  residual  aerosol not  col-
lected by the  control device.   This can be   addressed  by analyzing effluent
gas stream and ambient  meteorological  conditions using a  psychrometric chart
as shown in  Figure 5-2.!   The initial point is based  on the  measured stack
temperature and the average  gas moisture content  as estimated from  previous
stack tests.  The  final point  is the ambient wet bulb and dry bulb tempera-
tures.  -If- the line connecting the two  points  does not cross the saturation
curve,  then a steam plume probably is not present.  If the line crosses into

                                     10-4

-------
the saturation  region  and  then  reenters  the unsaturated region at a tempera-
ture of 5°  to  10° above actual dry  bulb  temperature,  then plume dissipation
probably occurs.
     The presence or absence  of  steam formation  within the plume can also be
evaluated by comparing  the dissipation of steam vent  effluent  at  the  plant,
with the  dissipation  of  the plume  of  interest.   If  the  steam  vent  plume
dissipates rapidly it  1s  unlikely that condensed water will  persist  long in
the contaminated plume.
     The color  of the  effluent is another  plume characteristic  which  should
be observed.   For fossil  fuel combustion  sources,  the color  is  an  indirect
indication of  operating conditions.   The following list  in  Table  10-1  was
compiled by EPA's Control  Programs Development Divisions:
	TABLE 10-1.  PLUME CHARACTERISTICS AND COMBUSTION PARAMETERS

Plume color	Possible operationg parameters  to  investigate	
White             Excess combustion air; loss of burner flame  in oil-fired
                  furnace.
Gray              Inadequate air supply or distribution.
Black             Lack of air; clogged or dirty burners or insufficient
                  atomizing pressure,  improper  oil  preheat;  improper  size
                  of coal.
Reddish brown     Excess furnace temperatures or excess air; burner con-
                  figuration.
Bluish white      High sulfur content in fuel.
For other types of  sources, the  color may  not be as variable or may not have
a distinct  meaning  with  respect to  the process  or the control  equipment.
Nevertheless, a  change  in the  color  indicates a  change  in  the  system.
10.2  BASELINE EVALUATION
     For diagnosing  the  change  in  performance  of  air  pollution  control
equipment, the magnitude  of  the shift in opacity  from  historical  (baseline)
levels is as important as the magnitude of the opacity itself.  A significant
increase in  opacity  usually  indicates an increase  in the  mass  concentration
of the effluent.  The  sensitivity  of opacity as a diagnostic tool  depends on
                                     10-5

-------
the characteristics of the opacity-mass concentration correlation.   Case 1 in
Figure 10-3  represents  the most  sensitive  relationship  between opacity  and
mass concentration  in that  a large  increase in  opacity  indicates a  small
increase in the mass concentration (and presumably the mass emission rates if
flow rates are similar).   In  Case 2,  opacity has diagnostic value  only up to
a certain  value  at which  point  it becomes totally  insensitive to  the  mass
                                  CASE 1
                    80
                                                 CASE  3
                   20
                                                  CASE 2
                            i       2       3      *
                            MASS EMISSIONS. ARBITRARY UNITS
                  FIGURE 10-3.  Opacity-Mass Relationships
                                    10-6

-------
emission rate.   Fortunately,  most  of the opacity-mass  concentration correla-
tions which have been  Identified appear  to  be  basically linear as represented
by Case 3 1n Figure 10-3.
     Figure 10-4 illustrates  a opacity-mass concentration  correlation devel-
oped at a coal-fired  power station  during  a five-year testing  program.   The
relationship 1s  typical  of  most  correlations  1n  that  it  is  approximately
linear.  It is  unusual  1n  that there 1s an abundance of test  data  with  which
to prepare the correlation.  Figures 10-5 and 10-6 present other correlations.
It 1s  apparent  that each  of  these  correlations have  a  different  slope  and
different degrees  of  variability.    For  these  reasons,  the correlations  must
presently be developed on  a  site-specific basis.  Procedures  for  calculating
the 95% confidence Interval for the  correlations are  presented 1n  Appendix F.
Usually 1t  1s  not  necessary  to  Include  the  confidence Interval  when  the
readings are used strictly for diagnostic purposes.
            0.30 ,
         J~ 0.25 -
          T: 0.20-
            0.15-
          ~ 0.10-
            o.os

Test

1
2
3
4
S

Symbol

O
A
Q
O
V
Date
0*t*
Taken
2/73
11/73
6/75
6/75
8/78

MW

?
r
140
80
125
. 50



- 40


. 30


. 20


• 10

 0
                     0.1    0.2    0.3 *   0.4     0.5    0.6    0.7    0.8
                              H - M«ss Concentration (g/N«J)
               FIGURE 10-4.  Example Opacity-Mass Relationship
                                    (from reference 2)
                                      10-7

-------
      0.2
     •0.1
                                                                        75  I
                                                                        70  5
                                                                           rtj
                                                                           
-------
10.3  LIMITS TO THE USE OF OPACITY
     Due to  the particle  size dependence  of light  scattering,  opacity  is
not universally applicable in evaluating mass concentration.  Particles above
approximately 2 microns  in  diameter  do  not  scatter light appreciably, there-
fore, substantial   mass  emissions  of these  relatively  large particles  may
occur without high opacity levels.  In such cases, the material  often settles
rapidly leaving deposits in.the vicinity of the discharge point.   Inspection/
maintenance personnel must  be  aware of  such sources and must utilize other
means to routinely evaluate control device performance.
10.4  REFERENCES
1.   Missen, R. and A. Stein, "Guidelines for Evaluation of Visible Emissions,"
     report by Pacific Environmental Sciences for the U.S.  EPA, Report No. EPA-
     340/1-75-007, April 1975.
2.   "Quality Assurance Guideline for Visible Emission Training Programs,"
     report by  PEDCo.  Environmental,  Inc.  for  the U.S.  EPA,  EPA  Contract
     No. 68-02-3431, Work Assignment III, August 1981.
                                     10-9

-------
                           11.0 SAFETY INFORMATION


      Inspection/maintenance personnel should not  be  working at an industrial
 facility unless they have the proper safety equipment, are trained in the use
 of this safety equipment,  and  are  trained  to  recognize potential  safety haz-

 ards.  All plant safety requirements and all agency safety procedures (in

 case of the regulatory agency Inspector) must be satisfied3.  This chapter em-
 phasizes the Importance of these safety procedures and provides  supplemental

 information concerning many of the most common procedures.

      There are a wide  variety  of potential hazards which may  be  encountered

 while inspecting and maintaining air pollution  control equipment.  A partial
 list of these hazards is provided below.
      1.   Acute  exposure  to toxic  gases such  as S02,  03,  N02>  ^S  and  CO
          due to  entry into  confined  areas and  due  to  fugitive  leaks  of
          flue gas  around  open measurement  ports,  corroded  door  gaskets,
          and corroded expansion joints  and  flanges.

      2.   Exposure  to fume  and dusts  containing  asbestos,  lead,  beryllium
          and arsenic.

      3.   Burns due to contact  with hot  objects such as uninsulated flue  gas
          ducts and portable sampling/ measurement  equipment.

      4.   Heat  stress due  to   exposure  to   hot  process equipment,  hot  flue
          gas ducts and stacks.

      5.   Static  electrical  shock  obtained  while  conducting   measurements
          downstream of electrostatic  precipitators or  in  fiberglass  ducts.

      6.   Falls  through   roofs  or  through   horizontal  surfaces  weakened  by
          the accumulation  of   excess  solids  or  the  corrosion  of  interior
          supports in air pollution control  equipment.


aThe information  and  recommendations  contained  in  this  chapter  have  been
 compiled from  sources  believed to  be  reliable  and  accurate.   No  warranty,
 guarantee or representation is made by  Engineering-Science  as  to  the absolute
 correctness or  sufficiency  of  any representation  contained  in  this  and
 other documents.  It  should not be assumed that all acceptable  safety measures
 are contained  in  this and other  publications  or that  other or  additional
 measures' may not  be  required  under  particular  or exceptional  conditions  or
 circumstances.   The  information provided in  this section is not  intended to
 super-cede, replace, or amend safety procedures  adopted  by  regulatory agencies
 and private concerns.

                                          11-1

-------
     7.  Falls on wet walking surfaces 1n the vicinity of control
         equipment or on the access ladders to control equipment.
     8.  Physical injuries due to entrapment in rotating mechanical
         equipment such as fan sheaves and screw conveyors.
     9.  Head  injuries  from  falling  objects,  overhead  beams,  and   pro-
         truding equipment.
     10.  Hearing impairment due to exposure to noise from rappers,
         compressors, and process equipment.
     11.  Explosions of fans operating at excessive tip speeds or
         operating severally out-of-balance.
     12.  Physical  injuries  due to  the improper  opening of  access hatches.
     13.  Asphyxiation due to improper entry into confined areas.
     14.  Asphyxiation  due  to   free  flowing  solids  discharged  from  hopper
         access hatches.
     15.  Eye injuries due to flying objects or splashing chemicals.
     16.  Foot injuries due to falling objects.
     17.  Explosions due  to static  electricity  discharge within  ducts being
         sampled or tested.
     18.  Exposure  to  toxic gases,  toxic particulate,  or steam  rising  from
         processes below  or present  in  the  general  vicinity  of  elevated
         walkways and platforms.
     19.  Burns due to contact with hot, free flowing solids.
     20.  Burns due to contact with high pressure steam leaks.
     Despite the long  list  of  potential  hazards,  inspection/maintenance work
on air pollution control  equipment need  not be dangerous.   Each individual
simply must develop respect  for these problems  and carefully  adhere  to the
safety procedures adopted by his or her employer.
11.1  GENERAL PROCEDURES
     Each individual  performing inspection/maintenance work  on  air pollution
control equipment should adhere to the following two basic rules:
     1.  Every situation should be  carefully evaluated  before work is begun.
     2.  Work should  be  halted if  the  individual  experiences  headache, eye
         or nose Irritation,  nausea,  dizziness,  drowsiness, vomiting,  loss

                                     11-2

-------
         of coordination,  chest pains  or  shortness  of  breath.   Conditions
         should be reevaluated before continuing the work.
     The first requirement can only be satisfied by conducting the inspection
or maintenance work  in  a methodical manner and  at  a  controlled pace.  Undue
haste can easily  result  in careless accidents.  It is also important to take
a rest break whenever fatigue reaches the point that judgement is affected.
     The second  rule is  necessary  because many of the  occupational hazards
encountered by  inspection/maintenance  personnel  do  not  have  any  easily
recognized characteristics  such as  a  distinctive  odor.   For  example,  some
chemicals such  as ^S,  which are  easily  recognized  at  low  concentrations
cannot be detected at  high concentrations  since at high  concentrations they
overwhelm and disable  the sensory organs.  Acute exposure  to  some materials
can occur  without any  immediate  discomfort,  then later (within  24 hours)
exposure will lead to  severe respiratory  Impairment.   For these  reasons  it
is important that all  personnel  be familiar  with the  initial   symptoms  of
acute exposure  to possible hazards  so that they are  able to leave  an area
of exposure  until  environmental conditions  can be more  fully  evaluated.
     Each regulatory agency and  industrial  firm should have  certain general
safety procedures to minimize the  risk  involved in equipment maintenance and
inspection.  These should,  at  a minimum, consist of the  following which are
covered in the next  sections:  Safety training, a medical  monitoring program,
and written safety procedures.
11.1.1  Safety Training
     All inspection/maintenance  personnel  should  receive  regularly scheduled
safety instruction in safety procedures,  use of  personal protective equipment,
and recognition of potential physical  and  inhalation  hazards.*»2   It 1s also
desirable that each  person receive training 1n First Aid and Cardio-Pulmonary
Resuscitation.2 Attendance at the safety training program should be mandatory
and should be  recorded  in each  individual's employment file.   New employees
assigned inspection/maintenance  duties should  receive  this training prior to
beginning field work.
     If the duties include  responding  to hazardous chemical  spills or fires
or the  Investigation of  chemical  waste  dump sites,  considerable additional
training may be necessary.  Information concerning safety procedures for such
work may be found 1n reference 1.
                                     11-3

-------
11.1.2  Medical  Monitoring Program
     Annual  physical  examinations  should  be conducted  as a  precondition to
further work.  These  examinations should provide  screening  for  evidence of
exposure to  toxic  substances.    They  need  not  be  a  comprehensive  health
evaluation since this  remains  the responsibility  of  each  Individual,  but
should, at a minimum, include the following tests:
     1.  Chest X-ray
     2.  Electro-cardiogram
     3.  Blood tests
     4.  Eye examination
     5.  Liver and kidney function tests
     6.  Pulmonary function tests
More detailed information is available in  reference 1.
11.1.3  Written Safety Procedures
     A written procedures manual should be prepared by  each regulatory agency
and private organization involved in the inspection/maintenance of air pollu-
tion control equipment.   This should at a minimum address use of personnel  pro-
tective equipment, recognition  of hazardous  conditions, recognition  of symp-
toms of exposure, procedures to  be followed in the event of potential  hazards,
and the reporting of  Illnesses or accidents.
     The manuals should be dated and revised whenever necessary.  Each emplo-
yee should be Issued  a numbered  copy of this manual and should be responsible
for adding modifications to the  manual.
11.2  COMMON INSPECTION/MAINTENANCE HAZARDS
     Safe working procedures begin with a  full  understanding  and  respect for
potential hazards.  Those problems most frequently encountered while working
on air pollution control  equipment are summarized in this section.  Additional
information is available in a number of excellent publications, including the
three listed below:
     1.  Occupational Diseases, A Guide to Their Recognition, Revised Edition.
         U.S. Department of Health, Education and Welfare, June 1977.
     2.  The Industrial Environment - Its Evaluation and Control, U.S.
         Department of Health, Education and Welfare, 1973.
     3.  Sax, N.I. Dangerous Properties of Industrial Materials, Van Nostrand
         Reinhold Company, New York, N.Y.
                                     11-4

-------
11.2.1  ^nhalatlon of Toxic Agents and Asphyxiants
     There are a wide variety  of toxic agents and  asphyxiants which may be ac-
cidently inhaled during the inspection of  an  industrial  facility.  The value
of a medical  surveillance program and proper safety procedures is demonstrat-
ed in the following paragraphs which summarize the potential  consequences of
acute exposure to some of the most common air contaminants.
     Soluble gases such as  sulfur dioxide, ammonia, and  chlorine have very dis-
tinctive odors and can often be detected by taste at very low concentrations.
At higher levels  eye,  nose,  and throat Irritation occurs.   Because  of these
"warning properties", these types  of air contaminants can usually  be  recog-
nized by  Inspection/maintenance  personnel  who  can  leave the immediate  area
or put on respirators.   If the  acute exposure  continues, these compounds can
cause severe cardio-pulmonary problems  Including but not limited to bronchitis
and pneumonia.
     Most organic compounds and  nitrogen dioxide are not very  soluble and these
often penetrate deep  into  the lower lung.  The initial  symptoms  of exposure
are nonspecific and may not be recognized by inspection/maintenance personnel.
These symptoms include:  dizziness,  drowsiness,  headache, 1ightheadedness, and
nausea.  The poor "warning properties" of these types of compounds makes them
extremely dangerous.  Acute exposure may  lead to pulmonary edema several hours
after the exposure.
     The chemical and physical  asphyxiants are  another group of  air contami-
nants which have very poor "warning  properties".  Chemical asphyxiants include
such gases as carbon monoxide  and hydrogen sulfide.  Carbon dioxide is the most
common physical  asphyxiant.   Inspection/maintenance personnel may  be  unable
to escape from confined and partially confined areas with high concentrations
of these compounds.
     The inhalation  of particulate matter  rarely causes immediate  physical
discomfort or  impairment.   For this reason,  it  is  possible for undesirable
quantities of  toxic  materials such  as  lead,  arsenic  and  asbestos  to reach
the lung.   These  can  be  slowly  absorbed  Into  the  blood system  thereby
allowing attack on  organs  such  as  the  liver  and kidney.   In  other cases, the
materials can  lead  to  severe  respiratory  problems.  Many   of  the problems
resulting from the Inhalation of particulate matter develop over an extended

                                     11-5

-------
 period  of  time  and  this  fact  underscores  the  need for a regular  occupational
 health  screening  physical.
      Table 11-1 is a brief summary of the initial symptoms (if any) of exposure
 to the  common  air  contaminants.   Since inspection/maintenance personnel  are
 usually more  at risk  from  short term, acute exposures than from chronic  expo-
 sures,  the acute symptoms are  emphasized.  Table 11-1 does not 11st all materi-
 als of  significant risk, does  mrt  11st  all  the possible  symptoms,  and does jiot
 provide a  complete  set of recommendations for personal  protection equipment.
 It 1s Intended  simply to emphasize the Importance of good safety procedures.
      The Occupational  Safety  and Health Standards regulations9 have  very spe-
 cific and  detailed  requirements  regarding the use of respirators for protec-
 tion from  exposure  to air contaminants.  These are  reproduced as Appendix D.
 Each individual  using  respirators  should ensure  that   the  requirements are
 satisfied. Basically this means  that each  individual  should  know how to
 select  the proper respirator, should  be  trained in the use  of  respirators,
 should  be  fitted  properly,  should  understand  respirator maintenance  require-
 ments,  and should  be physically  able  to  perform the  work   while  wearing a
 respirator.  Information  concerning  the use  of  respirators   is  available in
 reference  4,  pages   519 to   526 and  reference  7,  pages  1025  to  1055.
 11.2.2   Accidental  Falls
      Extreme  caution  is warranted  whenever  it  is  necessary to  walk across
 a  roof  or   walkway  which  has  a heavy accumulation  of  solids.  Occassionally
 these surfaces  have not  been designed  to support  this load plus  the   addi-
 tional  weight of  one  or two persons.   When crossing such areas it is prudent
 to  remain  close to the well trodden path.
      A  similar  problem may  be encountered  on  the roofs  of  air   pollution
 control  equipment.   In this  case,  corrosion  of the, interior supports can
 reduce  the  load bearing capacity  of  a roof, thus all should  be crossed  with
 caution.
     Around wet scrubbers  and any  other equipment using pumps there is   often
a layer  of. water 1n  the  adjacent walkways.   In  some  cases, this  layer of
water may  be  partially hidden by  a  cover of dust or  fibrous  material.  If
inspection/maintenance personnel  are  not careful  the  slick  area   will  be
discovered   in a most painful manner.
                                     11-6

-------
                         TABLE 11-1.
Compound
 Route of
   Entry
                           Symptoms of Exposure to Common Air Contaminants
                           (Source of data:   References 3, 5 and 6)
   Personal
  Protection
 Recommended
              corns
     Comments
Ammonia
Arsenic
Asbestos
Benzene
Beryllium
Inhalation
Inhalation
   and
Ingestlon of
dust and
fume

Inhalation
   of
airborne
fibers

Inhalation
of gas
Inhalation
of dust or
fume
Full face mask with
ammonia canister,
self-contained
rebreather or
supplied air
respirator

Self-contained
rebreather or
supplied air
respirator
Dust mask
Self-contained
rebreather or
supplied air
respirator

Self-contained
rebreather or
supplied air
respirator
Exposure may cause severe eye
and throat Irritation, nausea,
perspiration, vomiting and
chest pain.
Cough, chest pain, headache,
general weakness, followed by
gastrointestinal symptoms.
No symptoms,
Eye Irritation, headache,
dizziness, nauseai drowsiness,
convulsions.
Intense, brief exposure may
result In nonproductive
cough, low grade fever, chest
pains, and shortness of
breath.
Bronchitis or pneumonia
may result from an Intense
exposure.
Acute arsenical poisoning
due to Inhalation is rare.
Prolonged exposure may
cause cancer.
Acute exposure may lead to
unconsciousness and death.
Benzene 1s a confirmed
carcinogen.

Acute exposure to Beryllium
can result in chemical
pneumonitls with pulmonary
edema several hours after
exposure.

-------
TABLE 11-1.
Symptoms of Exposure to Common Air Contaminants  (Cont'd.)
(Source of data:   References  3,  5  and  6)
Compound
Cadmium
Carbon
Dioxide
Carbon
Monoxide
Ethylene
Oxide
Ethers
and Epoxy
Compounds
Hydrogen
Chloride
Route of
Entry
Inhalation
of dust and
fume
Inhalation
of gas
Inhalation
of gas
Inhalation
of gas
Inhalation
of gas
Inhalation
of mist or
vapor
Personal
Protection
Recommended
Dust mask
Self-contained
rebreather or
supplied air
respirator
Self-contained
rebreather or
supplied air
respirator
Full face
respirator and
protective
clothing
Self-contained
rebreather or
supplied air
respirator
Full face mask
Symptoms
Acute exposure may occur without
any Immediate symptoms. In
some cases there may be dryness
In the throat, headache, cough,
shortness of breath and vomiting.
Symptoms are those of asphyxia,
Including: headache, dizziness,
shortness of breath, weakness,
drowsiness, and ringing In ears.
Initial symptoms Include:
headache, dizziness, nausea,
drowsiness, pale skin color,
and vomiting.
Nausea, vomiting, eye and
nose Irritation, drowsiness.
Eye Irritation with tear
production, coughing, nausea,
drowsiness.
Eye and throat Irritation.
Comments
Brief exposures to very
high concentrations may
result In pulmonary
edema.
Rapid recovery occurs upon
return to fresh air.

Acute exposure may lead to
unconsciousness and pul-
monary edema.

Concentrations of 1,000 to
2,000 ppm are dangerous
even for a brief time.

-------
   Compound
                       TABLE 11-1.   Symptoms of Exposure  to  Common Air Contaminants  (Cont'd.)
                                    (Source of data:   References  3.  5  and  6)
 Route of
   Entry
   Personal
  Protection
 Recommended
           Symptoms
     Comments
Hydrogen Sulflde
Lead, Inorganic
Mercury, Inorganic
Nickel
Inhalation
of gas
Inhalation
of dust,
fume, or
vapor;
Ingestlon
Inhalation
of dust or
vapor
Inhalation
of dusts
Self-contained
rebreather or
supplied air
respirator
Self-contained
rebreather or
supplied air
respirator;
full body work
clothing
Self-contained
rebreather or
supplied air
respirator
Full face
respirator
Hydrogen sulflde has a
distinctive rotten eggs odor
at low concentrations, however,
at high levels there may be
no odor at all due to
olfactory paralysis.

Initial symptoms of acute
exposure Include:  headache,
dizziness, eye Irritation,
loss of coordination,
weakness, and numbness.

No Immediate physical symptoms.
No Immediate physical symptoms.
No Immediate physical symptoms.
Death may result rapidly
from acute exposure.
Chemical pneumonia may
develop several hours
after exposure.
Work cloths should be
washed separately from
street cloths.  Disposable
shoe covers may be advis-
able.  Lead Is a cumula-
tive poison.
Work clothes exposed
should not be washed or
stored with street clothes

Elemental mercury readily
volatilizes at room
temperature.

Nickel 1s a suspected
carcinogen.

-------
                       TABLE 11-1.
   Compound
 Route of
   Entry
                Symptoms of Exposure to Common Air Contaminants  (Cont'd.)
                (Source of data:  References 3, 5  and  6)
   Personal
  Protection
 Recommended
                           Symptoms
                                      Comments
Nickel CarbonyT
Nitrogen Oxides
  (NO. N02)
Ozone
Sulfur Dioxide
Sulfurlc Acid
Inhalation
of gas
Inhalation
of gas
Inhalation
of gas
Inhalation
of gas
Inhalation
of mist
Full face
respirator
Self-contained
rebreather or
supplied air
respirator
Self-contained
rebreather or
supplied air
respirator

Self-contained
rebreather or
supplied air
respirator
Full
mask
face gas
                Severe exposure can cause
                headache,  dizziness, nausea,
                vomiting,  fever,  and difficulty
                breathing.

                Initial  symptoms  Include:
                cough, chills,  fever, headache,
                nausea,  vomiting.
                Eye Irritation,  dryness  of  nose
                and throat,  choking,  cough,  and
                fatigue.
                Sulfur dioxide has  a strong.
                suffocating odor at levels
                greater than 0.5 ppm.   The
                taste threshold Is  0.3 to
                1.0 ppm.   Eye Irritation begins
                at approximately 20 ppm.
Eye Irritation, tickling of
the nose and throat, coughing,
and sneezing.  Concentrations
of 0.1 to 10 ppm may be un-
pleasant and concentrations of
10 to 20 ppm may be unbearable.
                                 Nickel carbonyl Is a
                                 confirmed carcinogen and
                                 may release CO upon
                                 decomposing.
                                    i
                                 Acute pulmonary edema may
                                 follow a five to twelve
                                 hour period with no
                                 apparent symptoms.  During
                                 exposure only mild bron-
                                 chial Irritation may be
                                 experienced.  Concentra-
                                 tions of 100-150 ppm are
                                 dangerous for periods of
                                 30 to 60 minutes.

                                 Pulmonary edema may occur
                                 several hours after severe
                                 exposure.
                                  Survivors  of acute
                                  exposure may later develop
                                  chemical bronchopneumonla
                                  Concentrations of 400 to
                                  500 ppm are considered
                                  Immediately dangerous to
                                  life.

-------
                       TABLE 11-1,
   Compound
 Route of
   Entry
                Symptoms of Exposure to Common Air Contaminants (Cont'd.)
                (Source of data:   References 3, 5 and 6)
   Personal
  Protection
 Recommended
           Symptoms
     Comments
Toluene
Vinyl Chloride
Inhalation
of vapors
Inhalation
of vapors
Chlorine
Inhalation
of vapors
Full face mask
Self-contained
rebreather or
supplied air
respirator
Full face mask
Eye Irritation, headache,
dizziness, fatigue, drowsiness,
loss of coordination.

Vinyl chloride has a pleasant,
ethereal odor; exposure may
cause llghtheadedness, nausea,
and symptoms similar to mild
alcohol Intoxication.

Pungent odor at greater than
3.0 ppm.  Throat Irritation
occurs at approximately 15 ppm.
Acute exposure may result In
cough, and chest pain, and eye
throat Irritation.
Acute exposure may lead
to collapse and coma.
Vinyl chloride Is a
confirmed carcinogen.
Pulmonary edema may occur
several hours after severe
exposure.

-------
      An  inevitable  chore  of inspecting  air  pollution control  equipment  is
 climbing up  and down  ladders.   During wet weather it is  probable that  the
 last person  to use  a  ladder deposited  a layer  of mud  on  the  foot   rungs.
 Climbing the ladder may be  dangerous  unless the  foot  rails  are used  for  the
 hands (rather than  the more common  practice  of grabbing the  side rungs).
 11.2.3  Noise
      Whenever the  noise level  1s  high enough such that  normal  conversation
 cannot be heard at  a  distance of  2 feet, hearing protection  1s necessary.
 Speech Interference  studies  have  indicated that  this  noise  level  is approxi-
 mately equal  to 82  to 88 dBA.7»8   For comparison purposes,  the OSHA limit
 for contluous exposure over  an 8 hour  period 1s 90 dBA.9
      During  Inspection/maintenance work,  noise levels greater than 90 dBA may
 be encountered around  electrostatic predpitator  rappers,  compressors  and
 blow tubes  for pulse  jet fabric  filters, Induced draft fans,  and  process
 equipment.   Even though the  Inspection/maintenance personnel  will usually not
 be 1n the high noise location  for  the  full 8  hour period, the  use of hearing
 protection is:  (1)   consistent   with plant  safety  policies,   (2)  ensures
 conformance  with  OSHA  requirements, and  (3)  may  improve  communication.  The
 latter point  may surprise some who  instinctively believe that communication is
 hindered by ear muffs or plugs.  Actually there is  some evidence to demonstrate
 improved hearing with  the  use  of  ear protection  when  noise levels exceed the
 90 dBA range.   It is believed that the hearing protection prevents distortion
 of the ear canal at the high noise levels while  1t keeps  the speech-to-noise
 levels at a constant level.8
      Failure  to use  hearing  protection can lead to gradual  hearing  loss es-
 pecially 1n the frequency range of 1000 to 4000 Hz.  Unfortunately, there are
 no  Immediate  symptoms  of  noise  Induced hearing loss.   The pain threshold for
 most  individuals is  130 dBA, well  above the 90 dBA level  which can adversely
 affect hearing,8
     There are  several basic  kinds of  hearing protection  including:  fixed
 size ear plugs, malleable  ear plugs, and earmuffsc  When used properly,  any of
these  can reduce levels of noise  30 to 50 dB  in the frequency  range  of 1000
to 4000  Hz.10*11  Obtaining  a proper fit with a tight accoustical  seal is

                                    11-12

-------
very important.  Some suggestions for obtaining the maximum degree of protec-

tion are listed below.

     (1)  If fixed size  plugs  are used,  the  fit  should be checked on a regular
          basis since the ear canal may increase in size with  regular use of
          any ear plug and since the ear plug may shrink slightly.

     (2)  Malleable plugs should  be  inserted before the inspector begins any
          work which  will  get his hands  dirty.   For this  reason,  the plugs
          should probably  be  inserted  immediately after  the  pre-inspection
          meeting (in the  case  of a regulatory  agency  inspector)  and remain
          in until the Inspection is over.  This will  prevent dirt or foreign
          objects from  causing   irritation  or  infection  of  the ear  canal.

     (3)  To obtain the  best  acoustical seal, the entrance  to the  ear canal
          should be  opened and  straightened by pulling  the  ear  backwards
          with one hand  while inserting the plug with the  other.   (For more
          information, refer  to  the  illustration  shown   in  Reference  7,
          page 333).

     If adequate noise  protection  cannot  be obtained by the use  of  a single
type of protector, then an inspector may wish to use an ear muff in combination

with a plug.  This provides an additional attenuation of 3 to  5  dB  over that
available with either one alone.10

11.2.4  Head Protection

     Inspection/maintenance-personnel should wear hardhats  whenever  they are
on plant  grounds  (except  of  course  when inside  office  spaces  and  similar

sheltered areas).   The  hardhats  should  satisfy the  requirements  of  ANSI
Standards for  Protective  Helmets,   Z89.1-1969.   These  should  protect  the

individual from falling  objects  (impact force less than 850 Ib-ft)  and from
collisions with low crossbeams or small equipment suspended  from the ceiling.

     At least once a  month, the  hard hat should be checked for possible cracks

or a defective suspension.  The latter is particularily  important since it is
the suspension which  provides  the impact protection.  Instructions for regular

washing of hardhats is provided in Reference 12.

11.2.5  Eye Protection

     Inspection/maintenance personnel routinely  pass through process equipment

areas where eye protection  is  mandatory.  Glasses with side  shields or flexible

goggles should be carried and  put on when entering these  areas.   The eye pro-
                                    11-13

-------
tection must conform to the requirements of ANSI Standards Z87.1-1968.  Prac-
tice for Occupational  and  Educational  Eye and Face Protection.  Contact lenses
should not be worn.
11.2.6  Footwear
     Many Industrial facilities do not  require  safety shoes as long as entry
to certain areas 1s not necessary.  Nevertheless, safety shoes should be worn
at all times while  performing  Inspections/maintenance.   Shoes  with metal toe
boxes are  generally adequate.   The  footwear  should  meet  the  specifications
of ANSI Standard Z41.1-1967.
11.2.7  Burns
     Most of the  gas  streams  of  concern  to air pollution  control equipment
Inspection/maintenance personnel  are at relatively  high temperatures.   For
example, the gas streams from cupolas  and stoker fired boilers are often 400°F
to 600°F.  The ducts transporting this effluent gas  stream are often uninsu-
lated and contact with a duct surface can  result 1n a burn.
     Handling pi tot tubes  and  other sampling  equipment  recently  withdrawn
from a  hot  duct  is another common  cause  of  burns.   Insulated  gloves  are
available for this  purpose.
11.2.8  Heat Stress
     Heat stress  is a  potential  problem  for  regulatory  agency  inspectors
since they do not  have  the opportunity to become  acclimated  to  hot environ-
ments.  Measurements  of  effluent  conditions  often  brings  inspectors  into
close proximity of  hot  ducts  and/or  stacks, both of  which radiate heat.   If
preventive measures are not taken, either  heat  cramps or heat exhaustion may
result.  The symptoms of  these problems are severe  cramps,  extreme fatigue,
weakness, nausea,  clammy  or moist  skin,   high  pulse rate,  and  a  high  body
core temperature.13  To reduce  the  risk of heat stress  the  inspection/main-
tenance personnel  should take the precautions  outlined below:
     (1)  Take frequent  breaks  1n a cooler  area.
     (2)  Drink  fluids recommended by a physician or Industrial
          hygiene professional  for such conditions.
     (3)  If possible, schedule strenuous  work  and  hot work in
          the cooler parts of the day.

                                    11-14

-------
11.2.9  Cold Weather Conditions
     Routine inspection of air pollution control equipment should not be done
when the wind chill factor is below -20°F.  The wind chill  factor is a combined
Index of the ambient temperature and the wind  velocity as  shown in Table 11-2.
Below -20°F there  is an Increased risk  for localized cold injuries.  There is
also a reasonable chance that measurements made under the stress of excessively
cold conditions  win not  be accurate due either to the  impaired  judgement of
the Inspector or due to the malfunction of portable instruments.   Additional-
ly, during  such periods,  plant  maintenance  personnel  often must cope with
process operational problems caused by  the cold conditions.  Thus, regulatory
agency inspectors  should  not  perform any Inspections  in very cold  weather
unless warranted by  a  public  health emergency or  a  severe  problem demanding
immediate attention.
11.2.10  Electrical Shock
     High static electrical charges can build up 1n certain ducts  and control
systems.  When  Inspection/maintenance  personnel  attempt to  insert  a  pi tot
tube, thermocouple, or other probe into the duct they may receive  a consider-
able shock.   In some cases,  the static  discharge could initiate a  serious
explosion.  Before  inserting   any  portable test  equipment  or  cleanout  rod
Into a  duct and  gas stream,  the  test equipment  probe and  rod should  be
grounded.*  This consists  of:   (1) a  proper electrical bond between the  probe/
rod and the sidewall of the duct (or  control  device)  and (2)-a  proper  ground
connected to the sidewall.
     The existence of a proper ground for the sidewall must be confirmed with
the appropriate  plant  personnel.  If  1t 1s  impossible  to prepare a  proper
bonding wire £r 1t is impossible to  confirm adequacy  of the  sidewall  ground,
the measurement should not be conducted.
     Also before a measurement is  conducted,  1t is important to  confirm that
the proper grounding equipment  has been  selected and that  there are no defects
which would interfere with the static electricity dissipation.   Each  bonding
wire and  ground connection should  be  visually  and manually checked  before
the measurement is made.
*For additionalfnformation on  grounding see National Fire Protection  Associa-
tion Publication No.  77  (1983)  and  American  Petroleom  Institute  Recommended
Practice 2003,  Fourth Edition  (1982).
                                    11-15

-------
TABLE 11-2.  Wind Chill  Factors
Wind in
ffl.p.h.
Calm
5
10
15
20
25
30
35
4o
>*5
50

32
Local Temoerature In Oeqress Fahrenheit
23
14
5
-4
-13
-22 •
"-31
Eouivalent Temoerature (Wind Plus Local Temoerature)
32
29
18
n
7
3"
1
-1
-3
-3
-4
23
20
7
-1
-6
-10
-13
-15
•17
-18
-18
Little— >
Danger
for Those
Properly
Clothed
14
10
-4
-13
-19
-24
-27
-29
-31
-32
-33
5
1
-15
-25
-32
-37
-41
-43
-45
-46
-47
-4
.a
-26
-37
-44
-50
-54
-57
-59
-61
-62
Considerable ^
Danger
-13
-18
-37
-49
-57
-64
-68
-71
-74
-75
-76
-22
-28
-48
-61
-70
-77
-82
-85
-87
-89
-91
-31
-37
-59
-73
-83
-90
-97
-99
-102
-104
-105
-40
-40
-47
i
-70 |
-88
-98
-104
-109
-113
-116
-118
-120
Extreme ^*
Danger
            11-16

-------
     The portable  instruments  should not be taken into plants and facilities
where there  is  a chance  of  exposure to explosive gases,  vapors,  fibers, or
dusts.  Such  industrial  operations  include,  but are  not  limited  to:   grain
elevators, grain milling  operations, coal  dust handling facilities,  and many
types of chemical  processes.   The battery operated thermocouples, pH meters,
and typical flashlights are not explosion proof.
     A regulatory  agency  Inspector should never  attempt to  open  an  electri-
cal cabinet  to  make  ammeter  readings.  Such  measurements should always  be
made by qualified plant personnel.
     Inspection/maintenance work  should not be  conducted  during  threatening
weather conditions.   Lightening  sometimes  travels  a   considerable  distance
ahead of an  approaching  storm  and normally  strikes the high  areas of a plant
where the air pollution control equipment is usually located.
11.2.11  Ionizing Radiation
     The new hopper  level  detection  instrumentation  used  frequently  on large
scale fabric filter and electrostatic precipitator systems  employs a source of
gamma radiation to detect the accumulation of solids.  If the source  unit has
been damaged due to  impact from moving  equipment  or  some  other cause,  exces-
sive radiation  levels  in  the immediate vicinity  of  the source could  result.
Normal safety  procedures  in  this  case include  posting   warning  signs  and
roping off the  damaged area  until the  source  can be sealed.   The regulatory
agency inspector should thus be cautioned never to approach a damaged nuclear
type hopper level detector.
     When entering the  hopper,  plant maintenance personnel  should  carefully
complete the lock-out  procedure  for the nuclear type level  detector.   The
shield must  cover  the  source  unit  in  order  to  provide  protection.
11.3  CONFINED AREA ENTRY
     A confined  area  may be defined as an  enclosure  in which dangerous  air
contamination cannot  be  prevented or  removed  by  natural  ventilation  through
openings.   Access  to the  enclosure  1s  usually  restricted,  so  that  it  is
difficult for a worker to escape or to be rescued.  Examples  of such  confined
areas Include  fabric filter  compartments,  electrostatic  precipitator pent-
houses and electrode  sections,  wet scrubber internals, mechanical  collector

                                    11-17

-------
  internals, and  hoppers  on  all  types  of  air  pollution   control  equipment.
  Maintenance personnel  must  enter these  areas on  regular  basis;  regulatory
  agency inspectors  should  not  enter  confined  areas.   It  is  important to
  note:
           Plant maintenance  personnel  should be  trained   in  proper  confined
           area entry  procedures,  should be  given the  necessary respirators
           and personal protective equipment,  and  should  be required to  per-
           form all  work  in strict  adherence  to plant  safety  procedures and
           OSHA requirements.
      Potential dangers of confined area entry can be grouped into three cate-
  gories:  oxygen deficiency, explosion, and exposure to toxic air contaminants.
  Oxygen deficiency is a very common and insidious hazard since many of the gas
  streams treated 1n air pollution control equipment have oxygen concentrations
  of only 4% to 12%.   Oxygen  levels of less than 19.5% produce detectable physi-
  ological changes, while levels less than 16.0% result 1n rapid disability and
  death.  Unfortunately, the initial  sensory symptoms  of  low oxygen levels are
  simply a slight difficulty in breathing and ringing in the ears.  Maintenance
  personnel may  ignore  these symptoms  and  then rapidly  become  incapacitated.
  Prior to entering any  confined  space, the environment inside  should  first be
  tested using  a  portable oxygen  analyzer.   The probe  should be  long  enough
  so that the  gas  actually _[n_ the confined area  is tested  rather than the air
  leaking in around the partially opened hatch.
      Maintenance personnel  must be cognizant of the explosion potential  inside
  air pollution control equipment.  Many combustion sources  generate moderately
  high concentrations  of carbon monoxide, especially during  upset periods.  A
  combustible gas analyzer  should be  used  to  detect  the  presence  of  CO and
  organic vapors.  The particulate matter collected in  many control devices is
  also potentially explosive.
      Some of the most  common  toxic  air  contaminants  1n  the confined  areas of
  air pollution control  equipment are  ^S, 303, N02 and particulate. Maintenance
  personnel  should select  and use the  proper  respirator  based  on the initial
 sampling of the interior environment of each  confined area.*
*The selection and use of respirators are subjects too involved and too impor-
 tant  to be" covered  briefly;  for these reasons, they are  beyond the  scope of
 this  manual.   For detailed  information please read the appropriate references
 listed  earlier.
                                     11-18

-------
11.3.1  general Entry Procedures
     The National  Institute of  Occupational   Safety  and Health  (NIOSH)  has
presented specific  guidelines  for working  inside confined  areas.   Excerpts
from this document are provided  in Appendix  E.  Maintenance personnel  should
study the excerpts  in Appendix  E and  then  read the entire  NIOSH document.
     A written permit should be  obtained prior to entry into confined areas.
This should confirm that  the'establlshed  lockout and isolation procedures have
     •
been followed and the personnel have the  proper safety equipment.  This permit
should be posted in  a conspicuous location.  During the lockout the air pollu-
tion control  system and all accessory  systems should be locked  and  tagged.
The accessory systems include:  fans, rapper systems, conveyors, compressors,
solid discharge valves, hopper and Insulator heaters, hopper level  indicators,
and pumps.   If  more than  one  person  1s  involved  1n the entry,  each  person
should have his  own lock and key for  each  Hern which must be  locked  out.
     The access  hatch to the  confined area  should  be  opened with  extreme
caution.  If  a  unit which  operates   at  a positive  pressure  has  not  been
deenergized yet, the hatch  will  have as  much  as  300 pounds of force  on  the
interior surface of  the  hatch (6 inches  static pressure, hatch  2'x4').   Once
the last bolt or  latch  has been removed, the  hatch  will accelerate  outward.
Units which operate  at  negative  pressures have the opposite problem  in that
the hatch is held closed with several  hundred pounds  of force.   Personnel  who
wrap their hands around the hatch 1n a futile attempt to "break the seal"  can
suffer severe  injuries  1f  the  hatch   slams  shut again.   Opening of  hopper
access doors must be done with particular care since  hot, free  flowing solids
may be piled  up  behind  the door.  Once the hatch is  opened  these  solids  can
suffocate, severely  burn,  and/or otherwise  injure personnel  below the access
hatch.  Additionally, many  of  the  solids collected  1n  air  pollution  control
systems are combustible.   When the hopper access  hatch  is  removed the Inrush
of ambient  air  (and  oxygen)  can cause  fires  inside  the  control device.
     Once safe access is  achieved, the environment should be carefully observ-
ed and tested.  The  oxygen  concentration and  the levels of potentially toxic
gases should be determined by portable meters.  The dust levels should be ob-
served.  Entry should not be made until the proper personal  protective equip-
                                    11-19

-------
ment has  been put on  and  checked out.   There  should always  be  an observer
1n the  vicinity of the  access  hatch who  1s  trained  and  equipped  to rescue
the worker(s).
     There are  several specific confined area entry problems for each type of
air pollution  control  system.   These are  summarized  in the  following sec-
tions.
11.3.1.1  Electrostatic Precipitators -  A predpltator which has been deener-
gized for a period up to several days may still  have a residual electrostatic
charge  of 10 to 20 kilovolts on the electrodes.   The electrodes could deliver
a  fatal  shock  unless  this  charge is  first dissipated to ground using a "hot
stick".  A "hot stick" normally  consists  of a long  wooden  stick with.a metal
hook on  one  end and a flexible metal  conductor  connected  to  a ground.  This
should  be used  each time  entry  is made into  a penthouse and  electrode zone.
Regular maintenance (primarily  involving  cleaning and testing) is  necessary
for the  hot  sticks.   If the sound  of "buzzing" characteristic of  a  hive of
bees is  ever heard upon  opening of  a hatch, the  unit  is still  energized.
Work should be  immediately halted!
     While working inside  an electrostatic precipitator, a  securly  connected
ground  should be kept in  plain view at all time.   If the power supply is acci-
dentally energized this  ground  could provide some  degree of  protection.
11.3.1.2  Fabric Filters - Large fabric filter systems are  often built with a
number of compartments, one or-irore of which may be Isolated for maintenance.
Entry to the  off-line  compartments  for maintenance  should be done with the
same care used  for entry  Into  any  other  confined  area.  The  dampers  which
isolate one  compartment  from another often  leak  which can  result  in  high
concentrations of toxic gases such  as ^S  and  CO  in the off-line compartment.
     While performing  maintenance  work Inside a  fabric  filter,  care should
be taken that  the self-contained  rebreather or  supplied  air  respirator do
not damage the  bags adjacent  to the  walkways;  there is  often  very little
clearance inside fabric  filters.   Before  entering  fabric  filters,  the  size
of the hatch should be considered  with respect  to the difficulty in removing
a  worker in  an emergency   situation.  Maintenance  personnel  inside  fabric
filters  (also electrostatic preclpitators and mechanical  collectors)  are

                                    11-20

-------
 Particularly vulnerable to  heat stress due  to  the additional work  inherent

 in the use  of  respirator  and the moderately high  temperatures  usually  found
 inside this equipment.

      The purpose of the preceding discussions is to  emphasize the importance of
 proper entry  procedures  for  air pollution  control  equipment.   Some  basic

 considerations have been introduced.   Individuals involved in this work should

 carefully read  OSHA  requirements9  and  other  references  on  the  subject.


 11.4  REFERENCES

 1.  Safety Manual  for  Hazardous  Waste Site Investigation,  U.S.  Environmental
     Protection Agency.  Draft  Manual  dated  September 1979.   Available  from
     the Office of Occupational  Health  and Safety, Waterside Mall, 401 M Street
     S.W., Washington, D.C.

 2.  Stack Sampling  Safety  Manual,  Prepared by Norman V. Steere  &  Associates
     for the U.S.  Environmental  Protection  Agency,  Contract   No.  68-02-2892,
     September 1978.

 3.  Occupational Diseases, A Guide to their Recognition,  Revised  Edition,  U.S.
     Department of Health,  Education and Welfare, June 1977.

 4.  The Industrial Environment - Its Evaluation and Control,   U.S.  Department
     of Health, Education  and Welfare, 1973.

 5.  Dangerous Properties  of Industrial Materials,  Sax,   N.I.    Van  Nostrand
     Reinhold Company, New  York, N.Y.

 6.  Accident Prevention Manual for Industrial Operations, National  Safety
     Council, 6th Edition,  1971.

 7.  Patty's Industrial Hygiene and Toxicology,  Third Revised  Edition, Volume  1.
     G. Clayton and F. Clayton, Editors.Wiley-Interscience Publication  Page  340,

 8.  Reference 6, Page 1152.

 9.  Occupational  Safety   and  Health  Standards,  U.S. Department  of  Labor,
     Federal  Register,  Vol.  37,  Number 202,  page   22157,  October  18,  1972.

10.  Reference 6, Page 1155.

11.  Reference 7, Pages 337-338.

12.  Reference 6, Page 1149.
           ;
13.  Reference, Pages 937-948.

14.  A Primer  on Confined  Area  Entry,  Rexnord  Safety Products,  Malvern,  Pa.


                                     11-21

-------
         12.0  ADMINISTRATIVE AND LEGAL ASPECTS OF PLANT INSPECTIONS

12.1  INTRODUCTION
     The majority of this inspection manual deals with Counter-flow Inspection
Techniques for the  actual  performance  evaluation  of  air  pollution  control
equipment.  An equally  important consideration  in conducting  air  pollution
field inspections is  the administrative  aspects  of the inspection  including
preparation, entry to the plant, post-inspection procedures, and report writ-
ing.  The legal  aspects  and consequences  of the  enforcement  inspection must
be considered prior to visiting any facility.
     This chapter presents the administrative and legal aspects of conducting
a field inspection.  Due  to the complex  nature of  environmental  law and the
differences in statutes from state to state it can not be an exhaustive study
of the subject,  nor can  it  be  used  as the final word  on the specific laws or
regulations.  Rather, the aim  of this chapter  is  to bring to the prospective
inspector's attention the  various  administrative  elements  of  the entire in-
spection scheme  which  are essential to a  successful and  complete  regulatory
inspection.  It  covers  procedures  for  preparation for an  inspection, plant
entry practices,  points to  be  aware of during the  inspection  and  some post-
inspection and  report  preparation  considerations.   In  addition,  special
sections of  this chapter  are  devoted  to methods  of  handling  confidential
business information and photographic documentation.
12.2  PREINSPECTION PROCEDURES
12.2.1  File Review   .
     Prior to  inspecting  a  facility  an  agency  inspector  should review the
agency files  concerning  that  facility  to  thoroughly familiarize  himself
with all  pertinent  information on the  facility's particular process(es),
types of  emissions,  and  operating  history.   The  following items  should be
checked.  Copies  of  items 1 and 2  from  the  list  should  be obtained for the
inspector's  "working" file.

                                     12-1

-------
     1.  Pending compliance schedules and .variances,
     2.  Construction and/or operating permits pertaining
         to source processes,
     3.  Past conditions of noncompliance,
     4.  Malfunctions reports, and
     5.  History of-abnormal operations.
The  inspector  should  also  obtain a copy of appropriate plant layout drawings
to assist in locating emission points,  drawing flow  diagrams, and in preparing
the  inspection  report.   If available in the  files,  the  inspector should note
what personal  safety  equipment  is required.   The files should  always  be re-
viewed before entry to the plant so that important plant characteristics will
be more  easily  remembered.
     The inspector  should  prepare a concise  "working"  file  containing basic
plant information, process descriptions, flowsheets, and acceptable operating
conditions  (Appendices A and B).   It should contain the following to facili-
tate inspections and/or preparations:
     1.  A chronology of control actions, inspections, and
         complaints concerning each major source in the plant;
     2.  A flowsheet identifying sources, control devices,
         monitors, and other information of interest;
     3.  The most recent permits for each major source;
     4.  Previous inspection checklists; and
     5.  Baseline performance data.
Once compiled,  such a file is easy to update and allows an inspector to do an
effective file  review without  struggling through  the typically  voluminous
central  files.
     The inspector should know what regulations are  applicable to the facility
and if he has any questions should consult the agency administration or legal
staff prior to conducting the  inspection.  It is  important  to know what  data is
useful  in confirming compliance  and what evidence is  necessary to document pos-
sible violations of each of the applicable regulations.
     Based on reviews of agency and any previous "working" files, the  inspec-
tor should schedule an inspection time when plant processes will probably be
                                     12-2

-------
operating at  representative  conditions.   The scheduling  of a time  to  visit
plants having batch operations or  other  irregular  operating schedules (e.g.,

seasonal) is especially important.

12.2.2  Inspection Announcement

     Agency supervisory personnel  should  be consulted concerning agency policy
on the advance announcement  of Inspections.   If 1t is  desirable  to  announce

the Inspection 1n advance, leads of 1 day to 1 week are generally adequate to
ensure that  the  necessary  plant   personnel  will   be  available.  The  person

contacted should be one  having the authority to  release data and samples and

to arrange for access to specific processes.

12.2.3  Inspection Equipment

     Necessary tools  and  safety  gear should  be   organized  and  be  ready  to

carry in a portable case from emission point to emission point.

                   Carry at all Times

                   Hardhat
                   Safety glasses or goggles
                   Gloves
                   Coveralls
                   Safety shoes (steel toed)
                   Ear protectors
                   Dust masks
                   Tool belt
                   Tape measure
                   Flash! ight
                   Differential pressure gauges
                   Gauge magnets
                   Thermocouples
                   Multimeter  and  grounding  cables
                   pH paper or pH meter
                   Duct tape
                   Pocket guide of Industrial hazards

Other equipment  should be  gathered to be left in  a  central  location such as

the inspector's  car until needed.

                   Carry When  Needed

                   Respirator with appropriate cartridge(s)
                   Ropes and harness
                   Stopwatch
                   Pryba r
                   Velometer
                   P1tot tubes
                   Bucket
                   Sample bottles  (Continued)

                                      12-3

-------
                   Combustion gas analyzer
                   Oxygen meter
                   Tachometer
                   Self-contained breathing  equipment
Particularly important is the  safety equipment—including the hard  hat,  the
safety glasses,  and  the  ear protectors.    Remember,  it  is  the  inspector's
responsibility to know what  safety equipment will  be  needed and to  have  it
before entering the plant.  "Access  to certain  industrial  facilities can  be
rightfully restricted or  refused by plant representatives if designated safety
equipment is not worn.
     Some plant may  require  that equipment  brought on  to plant property  be
signed in.  An equipment  inventory form may facilitate  check-in of the inspec-
tion equipment.
12.2.4  Plant Surroundings
     Observations of areas surrounding the plant before  entering  may  reveal  a
variety of  signs  of  operational  practices and pollutant emissions which  can
aid in the preentry evaluation, including:
     1.  Obvious vegetation  damage near the  plant,
     2.  Odors downwind of the plant,
     3.  Deposits on  cars parked closeby,
     4.  Other signs  of "dusting" downwind of the plant,
     5.  Fugitive emissions  near plant boundaries,
     6.  Conditions around the product and/or waste storage piles,
     7.  Conditions near lagoons and sludge  ponds,  and
     8.  Proximity of source to potential  receptors.
Some of the signs may indicate that  fugitive emission  sources should  be added
to the inspection agenda.  If odors  are a  problem,  the  weather conditions in-
cluding wind direction should be noted for inclusion in the inspection report;
once inside the  plant,  olfactory fatigue may (under certain  circumstances)
reduce the inspector's ability to detect odors.
12.2.5  Visible Emission  Observations
     In addition to observing the plant surroundings prior to  entry, the in-
spector can also perform visible emission  observations  at this time.   It is
                                     12-4

-------
possible that not all emission points will, be visible from a location outside
the plant property lines, but those that are may be  conveniently  read before
entry.  Chapter 10 details the procedures for these observations.
     It 1s appropriate for the inspector to inform plant management of excess
visible emissions  subsequent  to  their observation;  and at the same  time he
should find  out  what caused  them.   Notification  of  the plant officials  of
excess emissions gives them the opportunity to promptly evaluate the problem.
There may be statutes  which  require notification  and the  Inspector should
discuss this with his agency's legal staff.
12.2.6  Plant Entry
     Arrival at the facility must be during normal  working hours.   As soon as
the inspector arrives  on the premises he  should locate a  responsible  plant
official usually  the  plant  owner  or  manager.   In the  case of an announced
inspection this person would most probably be the official to whom notification
was made.  It is important that the inspector note the name and title of this
official. The inspector should be  prepared to  present the  official  with cre-
dentials and to summarize the reasons for  the  plant  visit.   Credentials pro-
vide the plant official with  assurance that  the inspector  is  a lawful  repre-
sentative of the  agency.  Proper credentials  should  include  the  inspector's
photograph,  signature, his physical  description,  (age,  height,  weight,  color
of hair and eyes), and the authority for the inspection.
     Unless  express permission has been  obtained from the inspector's employ-
er, visitor  release  forms and waivers  which purport to release  the  company
from tort liability should not  be signed.   An  Inspector may  waive his  right
to sue for  damages by  signing  these release  forms.   The precise  effect  of
signing an advance release of liability  for negligence  depends upon  the law
of the state  in which  the tort  occurs.   The State Attorney  General  can make
a current determination  of the  effect of  these waivers  for  each individual
state.  It 1s  suggested  that  the  inspector  consult  with the  legal  staff of
his agency  and  have  them obtain  a  determination.   If  the  plant  official
denies entry for refusal to sign a waiver, the Inspector should take down the
official's*name and  title,  leave the plant,  and contact his  supervisor for
further Instructions.
                                     12-5

-------
     Inspectors have, in the past, occasionally been asked to sign nondisclo-
sure statements  or agreements.   These  agreements  vary  slightly  in  content
from one  to  another, but  generally  require  that  confidential  information,
disclosed to an inspector during the course of  an inspection, be handled there-
after  in  a  specified manner.  An inspector should  not sign  such agreements
unless the  agency's  legal  counsel has  expressely approved the  agreement.
     To avoid subsequent legal entanglements,  an inspector must gain entry to
a plant's private premises and perform inspection activities in a manner con-
sistent with the company's  right of privacy guaranteed by the Fourth Amendment
of the U.S.  Constitution.   In a  recent Supreme Court  decision [Marshall v.
Barlow's. Inc. 436 U.S.  307 (1978)3, the  court  held that an  OSHA inspector
was not entitled to enter a work site without  either (1)  the owner's consent,
or (2) a warrant.   This decision has affected the policy for conducting inspec-
tions  under  EPA's  programs  and the state programs  and has made  it important
for all  federal  and state  inspectors to  conduct inspections  in  conformance
with their agency's policies.
     In order to have the background  information necessary to obtain a warrant
and following the guidance  of the Barlow decision,  there are several procedural
steps  that should be followed when an authorized plant official refuses entry
to an  inspector:
     1.  Tactfully discuss  the reason  for  denial  with  the plant  official.
         This is to  insure  that  it  has  not been  based on  a  misunderstanding
         of some sort.   If  resolution  is beyond the authority of the inspector
         he might  suggest  that the  official   seek  advice from the  company's
         attorneys on clarification   of EPA's or the  State's authority for
         inspection under the Clean Air Act and State law.
     2.  Note the facility name and  exact  address,  the  name  and  title of the
         plant officials approached,  the  authority  of the person issuing the
         denial (he must be authorized), the date  and time of denial, the rea-
         son for denial, facility appearance, and any reasonable suspicions
         as to why entry was refused.
     3.  The inspector  should  be very  careful  to avoid  any  situations that
         might be construed as threatening or  inflammatory.  Under no circum-
         stances should   he  cite  the potential  penalties  of entry  denial.
     4.  Lastly, the inspector should withdraw  from the  premises and contact
         his supervisor  to  decide on a subsequent course of action.
     These procedures also  apply in the additional case that consent is with-
drawn or partially withdrawn during  the  inspection.   In an  EPA "Memorandum

                                     12-6

-------
on Inspection Procedures"  (April  11.1979}1,  it  is  stated that all  evidence
obtained prior to the  withdrawal  of consent is  considered  admissable.   When
the inspector is denied access only to certain  parts  of  the  plant,  he  should
make note of  this  including  the area of the plant  and the  official's  reason
for denying access.  The  Inspection  should  be  completed  to  the extent  allow-
able and then, after  leaving the facility,  the Agency administration  should
be contacted  to  determine whether  a  warrant should  be  obtained  to  inspect
the portions of the facility not seen.
     All observations pertaining to  the  denial  should be carefully  noted  in
the inspection field  notes,  they  will  be   important  for  use in  affidavit
writing if a warrant is later sought.
12.2.6.1  Warrants -  In the  event  that a plant  official  persists  in  refusing
plant entry  or  withdraws  consent  during  the  course  of an  inspection,  an
administrative or criminal warrant may be used  to  gain entry into the  plant.
A warrant is  a judicial  authorization for an appropriate official to enter a
specifically described location and perform specifically  described inspection
functions.  The inspector  should always confer  with his  supervisor  to  deter-
mine that this course of action is the most  appropriate.
     There are two types of warrants:  criminal  and civil (or administrative)
warrants.  Administrative  inspection  warrants are the type most often  sought
the case  of  plant  entry denial; criminal  search warrants  are only used  in
cases where the inspection is intended, in whole  or in part, to  gather evidence
for a possible criminal  prosecution.   To obtain  a criminal search warrant, one
must be able to demonstrate criminal probable cause which is based on whether
a person  of  ordinary caution and  prudence  would be led  to  believe  and  con-
scientiously entertain  a   strong  suspicion   of  a violation.   Administrative
warrants are issued upon the showing  of (1)  civil probable  cause  or (2)  that
the establishment was selected  for inspection  pursuant to a neutral adminis-
trative inspection scheme.   Showing  civil  probable cause consists  of  demon-
strating specific evidence of an existing violation.  A  neutral or reasonable
Inspection scheme would include  schemes  such as annual inspections  of permit
holders in a given program.1
     When securing a warrant  three documents have to be drafted, an application
for a warrant, an accompanying affidavit and the warrant  itself.  The applica-
tion identifies the statutes and  regulations  under which  the  Agency  seeks the
                                     12-7

-------
warrant.   In  addition,  it  clearly  identifies  the  site or establishment to be
inspected  including,  if possible,  the name  of the owner  and/or  operator.
     The affidavit(s) in support of the  warrant application contain the fac-
tual background  for  seeking  the warrant.  An  affidavit  is  a sworn statement
signed  by  someone with personal knowledge of  the  facts stated  therein and
witnessed  by  a  notary public or the magistrate.   In  cases  where entry to a
plant has  been denied, at least one affidavit  would be signed by the inspector
who was  refused  entry.   An affidavit  for  a warrant sought  in the absence of
probable cause   should  incorporate  the  neutral  administrative  Inspection
scheme  which  was the basis  for  attempting to  Inspect  that particular esta-
blishment.
     The inspector works  in  conjunction  with the Agency  attorney to prepare
the warrant documents.  It is not important that the inspector be  able to word
his affidavit in proper legal terminology, but that he has carefully recorded
at the  time denial  of  entry, and obtained  previous  to  the inspection, the
pertinent  information needed to write the warrant application and affidavit:
     o  Address  of premises,
     o  Names  and titles  of  owner  and  authorized person who  refused entry,
     o  Time and place of refusal,
     o  Reason for refusal,
     o  Reason for inspection, and
     o  Description of premises and equipment to be inspected
It is important  that  the warrant specify to  the broadest extent possible the
areas that are intended to be inspected,  the records to be inspected, samples
to be taken,  and  articles  to be  seized.   In  general,  the stated  scope  of
inspection in  civil  warrants  can  be broader than in  criminal  warrants.
     When the agency attorney completes  the  preparation of the  legal docu-
ments, they must be  presented to the proper  judge  for his  signature  on the
warrant to  render it legally  binding.   For  an inspection  by EPA authority,
the warrant would be signed  by  a  U.S.  Attorney; state  inspection  warrants
are signed by a  state judge.   The assisting attorney should be able to direct
the state  inspector to the proper  court  and  judge.  The judge has the option
of questioning  the applicant  and  any   other  witnesses  before  deciding  to
issue the warrant.
     The conditions  of  serving a  warrant  vary from state  to  state and with
the federal courts, but some common things to be aware of are:

                                     12-8

-------
     o  Required  notice  to the  source .that  a  warrant  has  been  issued,
     o  Effective time  for  execution  and  return  of  warrant  to the  court,
     o  Reasonable time for inspection  (e.g.  during normal working hours), and
     o  Restrictions on forcible entry.
     When the inspector serves  the warrant  on  the facility owner or his autho-
rized representative, a copy of the warrant  should be  left with the plant offi-
cial.  The inspection must be conducted strictly  in accordance with the speci-
fications of the warrant.   The  inspector who conducts an illegal search can be
liable for civil damages.   If  samples  or records are taken,  receipts must be
issued.  The inspector  must maintain an inventory  of all  such  property taken
from the premises.
     When the inspection  has been completed,  the warrant is returned  to the
magistrate.  The person serving the warrant usually  signs the return  of ser-
vice.  A copy of the inventory of property removed from the facility premises
must be submitted to  the  court upon return of the warrant and  the  inspector
should be present to certify that the inventory is correct.
     There are three conditions under which a facility may be inspected with-
out the company  being  justified in  requiring  a  warrant.   The first  condition
would be in  an  emergency  when  there is  not enough time to obtain  a warrant.
Emergency situations include potential  imminent  hazard  situations  and  situa-
tions where  evidence  might  potentially  be destroyed  or  the violation  will
disappear before a  warrant  can be  obtained.   If the  facility  refuses entry
under these  situations, the inspector would  need  the assistance of a  state
or U.S.  Marshall to  gain  entry.   Secondly,  observations  by   inspectors  of
things that  are  in  plain  view, things that a member of the public  could be
in a  position  to observe,  also  do not  require  a  warrant.   This  includes
observations made from the public area  of a  plant or  from  private property
which is not normally  closed to the public.  Third,  some types of  abatement
programs specifically require right of warrantless entry.1
     It should be mentioned here  that  several special  instances  exist where
an inspector might  seek a warrant before  there has  been  any denial  of entry
to a facility.   These include instances when surprise is particularly crucial
to the  inspection,  when  a  company's  prior  refusals  or  other bad  conduct
Indicate it  will be  likely that entry  is  denied  again,  or when the distance
to a U.S.  Attorney  or  magistrate is  so considerable that  excessive  travel
time would be wasted if entry were refused.1
                                     12-9

-------
 12.2.7   Preinspection Interview
      Every  inspection  should begin with a  preinspection  interview.   The in-
 spector  should  plan  this  initial  interview with the plant  owner,  manager or
 other authorized  official  prior  to  the in-plant  inspection.   During this
 interview he  should  discuss the following points:
      1.  The  purpose and  scope of the inspection,                     ,..     »
      2.  The  type of measurements  to be made, and the type of samples  (tf any;
         to be  acquired,
      3.  The  operational status of all process and pollution control equipment
         to be  evaluated,
      4.  Inspector's preferred inspection agenda,
      5.  Changes in plant management that need to be  noted in the agency files,
      6.  Plant  safety requirements,
      7.  Scheduling  of a  post-inspection Interview,
      8.  Operating  records  required by Standards  of Performance   for  New
         Sources (NSPS)  and/or for  determinations  of  operating  conditions
         specified in permits,
      9.  Applicable  regulatory  requirements  and their  specific  applications,
     10.  Company's right  to  request  confidentiality of  business information,
     11.  Whether photographs are  allowed, and
     12.  Union  agreements.
      In  regard  to confidentiality, it is most important that the inspector be
 prepared to discuss how his agency or organization handles information claimed
 confidential  Including the procedures involved in determining confidentiality.
 A  suggested methodology concerning handling business confidential  information
 can  be found  in Section 12.5.
      Other issues the  Inspector should  be  prepared to  discuss in the  event
 that  company  officials  have  any  questions  concerning  them  include:
      1.  Authority for the Inspection,
      2.  Agency organization, and
      3.  Details of new or existing regulations.
 The inspector should request that a company official accompany him during the
 inspection to answer questions and to facilitate designation  of confidential
business data.
12.3  FIELD PROCEDURES
     The inspector should respect  all   union-company  agreements.   If these
agreements  so require, the Inspector should request  that union personnel make
any necessary measurements and  acquire  any samples.  This  1s done under the
supervision of  the  company  representative  and  with  the  guidance  of  the

                                    12-10

-------
 inspector.  The  assistance  of  union  personnel  considerably  accelerates  the
 field work since these  individuals  have  the  proper  tools,  and are thoroughly
 familiar with the  locations  of static pressure taps,  measurement  ports,  and
 control equipment  instruments.   The  inspector  should  not seek  information
 from union personnel  regarding the operation  and maintenance  at  the plant.
 This is unfair to both the union personnel and the plant representative since
 the union  operators  are not.  authorized  to release this  information  and  may
 not be  aware  of  the  confidential   nature  of   some   of  the  information.
      The inspector should  keep 1n mind  the  public  relations liason  part  of
 his role  as   inspector   and  seek  to  develop  or  improve the  good  working
 relationship with the plant officials.  An inspector  can  do this  by adhering
 to the following common sense rules:
      1.  Remain courteous at all times,  and
      2.  Respect the normal working schedule of the plant  personnel.
12.4  POST-INSPECTION PROCEDURES
12.4.1  Post-inspection Interview
      Having evaluated  the  exhaust  systems,  monitoring   equipment  control
 systems, and  possibly the  processes  themselves,  the  inspector should  again
 meet with a plant official  to:
      1.  Ask  follow-up questions as necessary,
      2.  Ensure that all confidential  Information has  been noted in the field
          data sheets,
      3.  Review  inspection  notes  so  that  there  is  general agreement  on  the
          technical  facts,
      4.  Discuss need for  a  follow-up inspection or  additional  records,  and
      5.  Prepare and deliver receipts  for any samples  taken.
      Under no circumstances  during these discussions  should the  inspector
 make judgements  or conclusions  concerning the compliance  of  the  source with
 applicable regulations.   He may state matters of  fact  (i.e.  smoke  opacity at
 a certain stack  exhaust  point was 20%), but should  not relate these to facility
 compliance.
12.4.2  File Update and Report  Preparation
      The source  files,  Including  Inspection  reports,  are  the  agency's source
 of Information  on  any  particular  facility, thus  the  inspectors  contribution
 1n updating  the files  and preparing  Inspection reports  is most vital.
                                     12-11

-------
      As soon  as  possible  after arriving .back  at  the office,  while all the
 events of the inspection are  still  fresh in his nrind. the  inspector should
 change all  appropriate  file  entries.   He  should  also  begin  preparing his
 inspection  report.   Most agencies have a  preferred  report  style,  so specific
 format will not be dealt with  in this document.   The report should, though,
 include (1) the conclusions of the inspector based on observations and cal-
 culations stated clearly 1n.concise paragraphs, (2) the  control device diag-
 nostic checklists found  1n Appendix D,  and  (3) a  cover page  (see example
 shown 1n Table  9) with the  following Information:
      -Any change 1n  responsible plant personnel,
      -Requested  permit changes or reported process modifications,
      -Results of the Counterflow evaluation,
      -Action  requested,
      -Inspector's signature, and
      -Date of Inspection.
 A copy of  the report  should  be kept both  1n the  inspector's "working" file
 and 1n the agency's  central file.
 12.5  HANDLING CONFIDENTIAL BUSINESS INFORMATION
      Industry is becoming  increasingly  sensitive to  and  aware  of  agency use
 and disclosure of confidential business information.  This subject has become
 a major point of contention  1n regard to allowing Inspectors on plant property.
 For this reason,  1t 1s especially  Important  for  all   agencies  performing
 Inspections to  adopt rules and  guidelines  for the handling  of confidential
 business information.  The presentation to companies  of an  organized, secure
 scheme for  confidential  information handling will do much  to increase their
 confidence  in an agency and their rapport  with Its representative plant inspec-
 tors.
     As  the current  time the Environmental  Protection Agency  and  most State
 agencies have established  policies  and  procedures  for handling  proprietary
 Information.  Most  of  these  include  penalties  for wrongful  disclosure and
 formal procedures  for safeguarding privileged  information.   Penalties  for
 Federal employees, for example, are fines up to $1,000 and one year in prison
 and dismissal  from employment  and  are   set  forth in  the  Code  of Federal
Regulations, 40  CFR  Chapter 1, Part 2,  Section 2.211 and the  United States
Code, Title 18,  Section  1905.   State  and local  agency employees should check
with agency attorneys to determine  their  personal  liability.   When state and
                                    12-12

-------
local agencies are delegated  authority  for the New  Source  Performance Stan-
dards and  National  Emission Standards  for Hazardous Air Pollutants  by EPA,
then they  may  be liable under the  same provisions  as  Federal  employees.
12.5.1  Defining Confidential Business Information
     From the inspector's standpoint, confidential information may be defined
as information received under a request  of confidentiality  which may concern
or relate to trade secrets,  processes, operations,  style of work or apparatus,
or the identity, confidential  statistical  data,  amount or  source  of any in-
come, profits, losses, or expenditures.   This information could be in written
form, in photographs, or in the inspector's memory.
12.5.2  Suggested Handling Procedures
     This section summarizes a suggested  set of  procedures  to be used by the
inspector and the  agency in  safeguarding confidential  information  obtained
during field inspections.  Although  each  inspector  and agency may use somewhat
different procedures  depending  on  the  agency policies, this  section should
serve to  bring  to  the  inspector's attention the  typical  methods   used  in
handling privileged information.
12.5.2.1  Receipt of Confidential  Business Data -  As  stated  previously, the
inspector must be  careful  to  identify  all  privileged  information collected
during an  inspection.   And,  since  confidential   information  involves extra
paperwork and possible  legal  consequences,  he should  avoid  collecting any
privileged information  unnecessarily.  Plant  officials should  be  given  an
opportunity to identify privileged information and  to request confidentiality.
They should  also  be  informed, if agency policy,   that  they may  also request
confidentiality at a  later  time  assuming  the  data has  not  already been dis-
closed.  The inspector  should  not,  in  any case,   speculate  whether  the data
in question will  eventually be considered  confidential; this determination is
a legal  and  administrative   policy  decision and  not within  the inspector's
authority.
     While the field inspector  is  on the  road,  inspection documents containing
privileged information should be marked and  kept in  a locked briefcase, while
physical samples should be  marked and kept  in a  locked container.  The com-
plete inventory  of  confidential  information  received  should  be  compiled  by

                                     12-13

-------
stamping each page  received  with an ink -stamp  similar to that  shown below.
At the  conclusion  of an  on-site inspection,  the inspector  should  prepare a
list of all documents received which indicates the following:
     (1)  A concise, general  description of each document,
     (2)  The total number of pages, and
     (3)  A document number.
                     CLAIMED CONFIDENTIAL
                DOCUMENT——  PAG
                REC'D. """"             OATE
                REC-D. "»v               PN
 12.5.2.2  Handling in the Office - Immediately upon return to the agency, the
 potentially confidential information (data, charts, drawings, etc.) should be
 placed in a secure, lockable file cabinet designated especially for confiden-
 tial  information.  Records should be kept of every person who uses a document;
 this  may be in the form  of  a  sign out  sheet for each document.  In addition,
 it may be useful  in keeping  document exposure to a minimum, that each document
 bear  an attached list of these persons  authorized to use it.
      In general, confidential  documents  should not  be reproduced.   If, for
 some  reason, it  becomes  necessary to copy privileged information,  all  copies
 should be included in  the confidential document inventory  and accounted for
 as would be an original.
 12.5.2.3  Privileged Data and Report Preparation  -  In preparing the inspec-
 tion  report, it  is recommended that confidential  information be reference in
 a non-confidential  manner  (i.e.,  by  reference to the  document in  the confi-
 dential files and a general  description of the information contained therein).
 If necessary,  the confidential  data may be  included in the  report,  but the
 entire report must  then be treated as a confidential document.
 12.5.2.4  Potentially Confidential Information  -  Although  all  inspections
 should include ample opportunity  for plant  officials to request confidential
handling of designated data,  every  allowance  should  be made  to  safeguard
potentially confidential  data  (it  may  be  stamped  "Confidential--Pending
Company Review") until  a company  Indicates  its intentions  or  submits a written
request.   In cases where an agency has questions  concerning the confidenti-
ality of certain information,  they may forward the information or inspection

                                    12-14

-------
 report  to  a responsible industry  official  for  company review.  The  official
 should  be  requested by cover letter to review the report and to clearly  mark
 any  data  considered confidential.   Until  the reviewed report is returned the
 data included  should  be treated  as  privileged information.
 12.6 USE  OF PHOTOGRAPHIC DOCUMENTATION
      Photographic  documentation  of an emission source, though useful in  some
 aspects, often elicits  a  negative  reaction from plant officials.  Therefore,
 1t is the  opinion  of  the  authors  that in most  cases the  inspection can be
 conducted  as  effectively and  usually  more efficiently  without the  use of
 photographic documentation.   There  will   be  some cases,  however,  where an
 Inspector  or his  agency will  require photographs of  control  devices and/or
 associated  equipment.   This  section  details  a   suggested  methodology  for
 photographic documentation and includes some particulars on the legal aspects
 of taking  photographs  of/within  the private areas  of a company's premises.
      Section 114  of the  Clean  Air  Act  may  be interpreted  to  say  that an
 inspector  can  take  photographs of  plant equipment to document his inspection.
 Like other  proprietary information,  EPA and other agencies can require that a
 source  allow photographs to  be  of  equipment that  might reveal  trade  secrets.
 But,  in almost all  instances it is best to avoid photographing any sensitive
 areas and to limit  photographs to  only those which  are absolutely  necessary.
 For  the purposes  of inspections  covered by this  manual,  the photographs  will
 generally be of air pollution control  equipment and any proprietary feature(s)
 in the  backgound can be shielded.
     Prior  to  taking  any photographs the inspector  should consult  with  the
 plant official (1) about plant policy concerning photographs and (2) to obtain
 consent.  This  is  mostly  conveniently accomplished during the pre-inspection
 interview.  The inspector may  offer to provide the  official  with  duplicates
 of all  photographs taken.   As  with other  business data  collected,  during
 and/or at  the  conclusion  of the inspection,  the  inspector should  ascertain
whether any of the photographs  taken contain proprietary  information and if
the company  wishes  to designate   any  as  confidential.   Photographs  taken
employing a Polaroid-type instant  camera  are useful  for  inspections because
they allow  an  immediate confidentiality  review  and the opportunity  for  the
inspector to readily provide the company with duplicate shots.
                                    12-15

-------
     The inspector must  be exacting  in -his photographic techniques  because
all photographs must  be  authenticated and thoroughly identified  if  they are
to be used as evidence in a court of law.   When a photograph is introduced as
evidence, the photographer, or  on occasion another person  familiar  with the
subject pictured, must affirm that the reproduction is a "fair representation"
of what  it  is trying  to show.   To  help  ensure  that  this  is  true  for his
photographs, the  inspector should take  Into  account  the  type  of  lenses,
filters and film  he  uses so as not to Introduce  any photographic artifacts.
He should also endeavor  to  take photographs that include some  indication of
(1) the scale of  the  object being photographed,  (2) its  location in the scheme
of the  plant  (if possible, without  including  proprietary  information),  and
(3) the direction from which the photograph was taken.
     Every photograph must be logged in the inspector's field notes.   This log
should include:
     o  Photo ID number,
     o  Type of film, lens, filters used,
     o  Date,
     o  Time,
     o  Location,
     o  Subject and purpose,
     o  Photographer's initials.
Polaroid-type instant  photos  should  be  immediately  identified  on  the  back
after shooting with  the   corresponding  photo  ID  number.   Photographs  which
require developing and printing should be  numbered  as soon after  these as
possible.  Photogaphs  of a  confidential   nature  must  be   developed  by  an
authorized contractor.
12.7  REFERENCE
1.  "Memorandum on Inspection Procedures (April 22, 1979)," from the  Assistant
    Administrator for  Enforcement  to  Regional  Administrators,  Surveillance
    and Analysis  Division Directors,  and Enforcement  Division Directors, U.S.
    Environmental  Protection Agency,  published  in The  Environment  Reporter,
    Bureau  of National Affairs,  Inc.,  Washington,  D.C.,  6-8-79,   pp  9-23.
                                      12-16

-------