EPA 340/1-87-002
Municipal Waste Combustion Systems
   Operation and Maintenance Study
                       Prepared by

                 Allen Consulting & Engineering
                    317 Howland Avenue
                      Gary NC 27513
                       Prepared for

                 EPA Project Officer: Louis Paley
            EPA Work Assignment Manager: Pamela Saunders
             U.S. ENVIRONMENTAL PROTECTION AGENCY
               Stationary Source Compliance Division
             Office of Air Quality Planning and Standards
                   Washington DC 20460
                       June 1987

-------
                                  DISCLAIMER
     This report was prepared by Allen Consulting and Engineering  for the Stationary
Source Compliance Division of the U.S. Environmental Protection Agency in fulfillment
of Purchase  Order No. 7W-6610-NASX.   The contents of this report are  reproduced
herein as received by the contractor.   The opinions, findings, and conclusions expressed
are those of  the authors and not necessarily those of the U.S. Environmental Protection
Agency.   The mention  of product names  does not constitute  endorsement by the U.S.
Environmental Protection Agency.
                                          11

-------
                               ACKNOWLEDGEMENT

      Appreciation is expressed  to the staff of  the Bureau of Sanitation,  City  of
Stamford,  and particularly to  Mr.  Marvin E.  Serra,  Superintendent  for  providing the
months of  operating  data and many hours  of assistance in helping us compile the field
study presented in Chapter 3.  In addition, appreciation is expressed for the assistance
provided by Peer Consultants, Inc. in preparing the statistical analyses performed on the
data generated by the field study.
                                        in

-------

-------
                                 TABLE OF CONTENTS
              LIST OF FIGURES.	„	 .v
              LIST OF TABLES	v

CHAPTER 1   INTRODUCTION	1

              1.1    Objectives and Scope of this Project	1
              1.2    Overview of MWC Sites Visited	1
              1.3    Report Organization	3

CHAPTER 2   GENERAL OPERATING CONCERNS	4

              2.1    Introduction	...	4
              2.2    Fuel Preparation and Handling	,	4
              2.3    Furnace  Operation	7
              2.4    Ash Handling	15
              2.5    Air Pollution Control Equipment	18
              2.6    Fans, Ducts, and Stacks	.. 24
              2.7    Startup and Shutdown	29
              2.8    Equipment Inspection Frequency	32
              2.9    Recordkeeping	34
              2.10   References	40

CHAPTER 3   SITE VISITS AND OPERATING VARIABLES STUDY	42

              3.1    Introduction	„.	 42
              3.2    Site Visits - Case Histories	42
              3.3    Operating Variables vs. Opacity Study	46
              3.4    References	59

APPENDIX A  AUDIT: OPACITY METER -STAMFORD MUNICIPAL INCINERATOR
                                       IV

-------
                                LIST OF FIGURES
Number
 2-1
 2-2
 2-3
 3-1
 3-2

 3-3
                                                          Page
Continuous Monitor Quarterly Report	•	35
Incinerator Shift Operations Record	37
Combustion Condition Observation Form	38
Schematic of Stamford Incinerator	48
Conceptual Relationship Between
Underfire/Overfire/Temperature	-51
Sample Data Sheet for Stamford Incinerator	«55
                                 LIST OF TABLES
 Number

  1-1
  2-1
  2-2
  2-3

  2-4

  3-1
                                                                         Page
 Types of MWC's Reviewed	2
 Typical Composition of Municipal Solid Waste	5
 Electrostatic Precipitator Log	22
 Typical Fan Inspection and Preventive
 Maintenance Schedule	•
 Example Incinerator Equipment
 Inspection Frequencies	
 ESP Parameters	
.26

.33
.53
                                         v

-------
                                    CHAPTER 1

                                  INTRODUCTION

             Municipal Waste Combustors (MWC's) have traditionally had problems in
     complying with participate  matter and visible emission  (VE) limitations  and are
     often the subject of citizen's complaints.  Due to stricter waste disposal regulations
     and decreasing availability  of  space for landfills, an  increase in incineration is
     anticipated.   It  is believed that, in many cases,  particulate matter and visible
     emissions  of  MWC's can  be reduced  by improving  the incinerator  and control
     equipment operation and maintenance (O&M) practices.
1.1  OBJECTIVES AND SCOPE OF THIS PROJECT
             The objectives of this project were twofold.   The first objective  was to
     determine the  significant operation and  maintenance considerations  that  bear
     directly on the day in/day out air pollution compliance status of MWC's.  Rather
     than  performing theoretical analyses of furnace and  air pollution control device
     performance,  achievement of this objective was sought by visiting several MWC
     facilities  and  discussing the  issues  on-site  with  operators  and  management
     personnel.
             The second objective was to determine  the effect of operating variables on
     visible  emissions  (opacity) at one facility.   Any correlation  found would provide
     insight  into the relationship between operating variables and opacity.

1.2  OVERVIEW OF MWC SITES VISITED
             An attempt  was  made  to select several facilities  which,  when taken
     together, would represent a reasonable cross-section of the types of equipment in
     service at the  time of the visits (1985 and 1986).   To keep  travel costs  to  a
     minimum for the  EPA headquarters and contractor's staff, site visits were limited
     to the eastern United States. Seven sites were visited with a total of 19 furnaces
     in operation.  Table 1-1 summarizes the types of MWC's reviewed.  Recent permit
     applications for construction of MWC's indicate  that the fraction  of operating
     MWC's  that burn  refuse derived  fuel (RDF) and/or utilize boilers to recover  heat
     will substantially increase in the future.  Processing municipal waste into RDF  is an
     operation that is  often performed off-site and then trucked to  the MWC facility.
     RDF processing facilities are outside the scope of this report.

-------
                                 TABLE 1-1



                         TYPES OF MWC's REVIEWED
Type
No. of of
Furnaces Furnace
4
2
3
1
4
2
1
2
Rotary
Combustor
Rectangular
Rectangular
Rectangular
Rotary
Combustor
Rectangular
Rectangular
Traveling
Grate
Continuous
or
Batch
Feed
Continuous
Batch
Continuous
Continuous
Continuous
Batch
Continuous
Continuous
Fuela
MB
MB
MB
MB
MB
MB
MB
RDF
Boiler
None
None
None
None
None
None
None
Water
Wall
Air Throughput
Pollution per
Control Furnace
Device (tons/day)
ESP
ESP
ESP
ESP
Venturi
Scrubbers
Baffle Wall
Scrubber
ESP
ESP
300
100
300
240
250
215
200
300
aMB = Mass burn (unprocessed municipal waste)



 RDF = Refuse Derived Fuel

-------
1.3  REPORT ORGANIZATION
             Chapter 2  of this report contains a discussion  of general O&M problems,
     concerns, and procedures described by MWC operators  during the site visits.  This
     discussion covers  the whole range of operations at  MWC's from furnace charging
     and combustion control to ash handling and air pollution  control equipment.
             Chapter 3  contains two major sections.  The first section describes eight
     brief case histories of specific O&M related air pollution problems and the solutions
     implemented by the MWC operators interviewed for this study. The second section
     describes the results of the study on  the effect of operating variables on opacity.
     This  discussion includes a description of the facility,  the  operating  variables
     studied,  and the analytical techniques employed.   The chapter ends with some
     conclusions that can be drawn of the study's results. FinaUy, Appendix A contains a
     report of the audit performed on the opacity meter at the facility.

-------

-------
                                    CHAPTER 2

                         GENERAL OPERATING CONCERNS

2.1  INTRODUCTION
             This chapter  discusses MWC operating concerns  as described by  MWC
     operators and their supervisors at seven facilities visited  over the course of this
     study.   The  operating concerns  discussed  herein  include operating practices,
     preventive  maintenance  requirements,  and  recordkeeping necessary to operate
     MWC's  reliably and in compliance with air  pollution emission limitations.  The
     operating concerns  are discussed  on a unit operation or equipment basis starting
     with fuel preparation and ending with the stack.

2.2  FUEL PREPARATION AND HANDLING
             Mass burn  MWC's are those facilities that charge  municipal  waste  to the
     furnace on  an as-received basis.  The alternative to  mass burning municipal waste
     is to  process the waste into refuse-derived fuel (RDF).  The extent of processing
     waste into  RDF varies  from  plant to plant, but usually it entails any one or
     combination of the following processes: magnetic separation, shredding, screening,
     and air classifying.  Most RDF facilities produce a fuel that can pass through a 10
     mesh screen (0.034 in. square), or  coarser fuels that can pass through  screens up to
     six inches square.   The  10 mesh  screen  produces a fuel  that can be fired in a
     tangentially fired furnace that is similar to  units that burn pulverized coal.  The
     coarser fuel is  fired  in  various types  of stokers,  most commonly  those with a
     traveling grate.
             The objective  of processing  municipal waste  into  RDF  is  to create a
     uniform  fuel by improving the fuel's  important combustion parameters, e.g. size
     distribution and combustible content.  Indeed, one of the major challenges faced by
     operators of mass burn furnaces is the wide swings in fuel quality that are inherent
     in unprocessed municipal waste.   This  subsection discusses  operating procedures
     employed   by  MWC operators  to deal with the difficulties imposed by burning
     unprocessed refuse  in mass-burn furnaces.

     2.2.1    Trash Composition
             Table  2-1 lists a "typical" composition of municipal solid waste.  Trash
     composition will vary among facilities depending on  factors such as region  of the
                                         4

-------
country,  season of the year, and the extent a given facility accepts non-hazardous
commercial and industrial waste.  Design values for heat and moisture content of
the waste will vary among facilities.  Factors that effect the design valves include
the nature of the waste anticipated for the specific facility and whether (and to
what extent) the waste will be processed into RDF.
                               TABLE 2-1

                       TYPICAL COMPOSITION OF
                        MUNICIPAL SOLID WASTE
        Component
Weight Percent
        Paper, Other than Newspaper
        Newspaper
        Garbage
        Yard Wastes
        Glass, Ceramic, Stone
        Metal
        Cardboard
        Wood
        Textiles
        Plastic Film
        Leather, Rubber, Molded Plastics
        Total
        Source:  Reference 1
     25
     14
     12
     10
     10
      8
      7
      7
      3
      2
      2,
     100%
        The combustion characteristics of any  given  truckload of waste  may be
substantially different than the design value or general composition guideline. For
example, municipal trucks collecting household garbage and trash will deliver waste
that  has substantially different  characteristics  than the  trucks for a specific
commercial operation that deliver  nothing but shredded paper or process  (non-
hazardous) waste from a plastic film or corrugated box manufacturer.  Regardless
of the source of waste, mass burn  MWC operators are often challenged by extended
periods of wet weather, when much of the material they are attempting to burn has
been soaked by several days of rain.

-------
 2.2.2   Pit Operation
        The fuel preparation and handling procedures discussed here apply directly
 to those  continuous and batch furnaces that utilize gravity feed chutes that are
 loaded with an overhead crane from a sub-elevation trash pit. A properly sized pit
 and one overhead crane can serve up to four furnaces with a total capacity of over
 1000 tons of refuse per day. Pit sizing is important to the efficient movement of
 raw  trash  through the  facility.   Considerations include  not  only  the  furnace
 capacity, but also the anticipated operating schedules for the furnace and the trash
 receiving operation.  Undersized pits require that municipal waste be  stored for
 significant  periods  of  time in a pile either inside the receiving room if space is
 available, or outside. In either case, odor problems are likely to ensue.
        Municipal waste  arrives  at  the incineration facility via  truck.  The trucks
 are weighed prior  to  tipping their trash  into  the pits.   Weigh records provide
.reasonably  accurate data regarding the amount  of trash processed over a week or
 more and such data is  useful in  establishing long term trends about the production
 of municipal waste for  the area serviced by the incinerator facility. The scale data
 can also  be used to  estimate  hourly  or  daily  furnace  throughputs,  but  these
 estimates are crude at  best since they do not account for changes in the amount of
 trash in the pit.
        The crane  operator can  make or break a successful MWC operation. The
 crane operator should  constantly  mix  wet and dry  refuse in  order to produce
 "uniform" refuse; although  he or she should keep some "wet" refuse in one area of
 the pit and some "dry" refuse in another area for emergency use when required by
 the furnace operator. Should it become necessary to go from feeding normal refuse
 to  feeding either wet or dry  refuse, the crane operator should do it gradually and
 give  the furnace  operator  time to change  the  combustion conditions.  Constant
 observation of incoming loads by the crane operator will give that person an idea of
 the condition of the storage pit at all times and  the location of the wet  and dry
 refuse.
        The feed hopper serves as a funnel to direct refuse into the feed  chute. To
 prevent bridging in many hopper  designs, the crane operator must take care not to
 fill the hopper more than approximately one foot above the entrance to the feed
 chute.
        When not feeding the incinerator units, the crane operator should utilize all
his time mixing the wet and dry  refuse in the pit.  Large non-combustibles such as
refrigerators, drive  trains,  etc. should be carefully removed from  the  pit.  Some

-------
2.3
facilities set such materials aside for a metal recovery contractor, others ship such
materials to  a landfill.   If  allowed to  pass  through the  furnace,  large  non-
combustibles  may damage the  grate system  and/or  the  bottom  ash handling
equipment.
        Similarly, certain  excessively combustible  items  such as  tires and large
foam rubber items should  be  set aside for disposal by  other means,  as should any
excessively large items such as chairs and couches.
        The  crane operator should  also attempt to clean the bottom of the pit
occasionally  and should not just take the refuse from the  top of the  pit.  The best
method is to remove everything down to  the bottom of the pit in one area, then
another, and so forth. If possible this program should be used weekly. The purpose
of periodically cleaning  the pit to the bottom  is twofold. First, the longer trash
stays in the pit, the larger its potential for creating odor problems. Second, decay
of municipal trash produces methane gas  that becomes trapped in the waste pile.
Thus, periodic turnover of the pile minimizes fire and explosion hazards.
        The  trash in the  feed chute provides  the seal between  the combustion
chamber and the pit room  for a continuous feed furnace.  If this seal is broken, the
combustion chamber will draw large volumes of  cold ambient air, thus upsetting the
combustion process and  creating significant air pollution  until the seal can be re-
established and  until the  furnace can return  to normal  operating  temperatures.
Therefore, the crane operator should never "get behind" a continuous  furnace.
        For  a batch furnace, when  the  furnace  operator determines that it  is
necessary  to charge the furnace,  a guillotine  gate located at  the bottom of the
charging chute is briefly  opened to allow the  trash to faU to  the furnace.  It is
important  that this procedure be  coordinated  with the crane operator to ensure
that trash is available in the chute and that the gate is open for the  minimum time
necessary so that as little  cold, ambient air as possible  is drawn into the furnace.
        Many facilities  maintain a backup crane so that the facility  can operate
when the primary crane is down for maintenance. In a  one crane operation, even as
little as one hour of crane  downtime may require initiation of shutdown procedures.

FURNACE OPERATION
        The  objective of  MWC  operation is to reduce the volume of municipal
 waste that eventually will go to landfill to the maximum extent practicable.  This
 objective is  achieved only when the MWC is  operated in a manner that provides
 good combustion.  Good combustion  requires  maximizing the  classic  three Ts of

-------
combustion; time,  temperature, and turbulence.  As  discussed below, the MWC
operator  has  meaningful  control of  only  temperature;  residence  time  and
turbulence having been maximized through the design process.
        This section  discusses operating practices  commonly utilized to provide
good combustion at MWC facilities.  Common upset conditions are described, as are
operating procedures  to minimize  the impact of such  upsets.  The reader should
keep in mind that providing good combustion minimizes air pollution by minimizing
the  emission  of unburned  particles and  maximizing  the combustion of volatile
organic compounds (VOC) that evolve from the municipal waste  in the combustion
chamber.
2.3.1   Combustion Air
        Oxygen  is  necessary  to complete  the combustion  reaction.   Oxygen is
introduced to the furnace as preheated or unpreheated ambient air. In order to
completely burn combustibles in the furnace, it is necessary  to provide combustion
air in  excess of the theoretical quantity required by stoichiometry to completely
burn (i.e., oxidize) the waste. A minimum of 50 percent excess air is necessary to
provide adequate turbulance of  the volatile gases  generated  from heating the solid
waste.  In general, waterwall furnaces require  excess air in the range of 50 to 100
percent excess air, while refractory furnaces require 150 to 200 percent excess air.
        An insufficient quantity of combustion air cannot provide adequate oxygen
and turbulence for good combustion. Such a condition produces increased emissions
of  carbon  monoxide,  unburned organic  gases and soot.   Curiously,  too  much
combustion  air also produces increases  in  soot, CO and organic emissions.  The
reason is that the  increased mass flow of cool combustion air  tends to  lower the
furnace temperature.  In addition, this increased  mass  flow  reduces the residence
time of combustible gases in the furnace and has  a tendency to entrain significant
quantities of ash out of the furnace.
        All  the  furnaces  observed in this  study  provide combustion air through
underfire systems and most also utilize  overfire  systems.  A brief description of
these systems follow.
        Underfire Air - Underfire air is provided by a fan with airflow controlled by
a fan inlet damper. The air is manifolded to several areas under the grates, where
it  is directed upward so that the air passes through the grates,  refuse and ash.  In
some MWC's the operator  can control the amount of air independently  to various
sections of the furnace.

-------
       Underfire air systems are designed to provide sufficient velocity (energy)
to the underfire air so that the air penetrates the fuel bed. This provides intimate
contact of combustion air with  the  solid,  combustible material in the fuel, thus
enhancing the combustion process.  However, in addition to  lowering the furnace
temperature, excessive flow of underfire air has the velocity necessary to carry ash
and  unburned solid  material out the  furnace in sufficient  quantities to create
visible emissions and particle fall out problems.
       Some incinerator systems utilize an air preheater to heat the underfire air
prior to introduction into the furnace.  A preheater is an air-to-air exchanger that
utilizes the waste  heat in the furnace exhaust gases.   Preheated underfire  air
increases the  flame  temperature,   thus  improving  the  combustion  and energy
efficiencies of the system.
        Overfire Air -  Overfire  air is introduced  above the fuel bed  via high
velocity  nozzles through the furnace sidewalls, perpendicular to the flow of fuel.
Optimum combustion  usually occurs when the overfire air  is somewhere  in  the
range of 40 to 60 percent of the total combustion air introduced into the system.
However, operating experience (i.e., trial and error) sometimes dictates that  the
fraction of overfire  necessary for  optimum combustion  will fall outside of this
range in a given furnace.  As for design purposes, overfire air is any air introduced
that  is not  underfire air.  Therefore, overfire air also includes air introduced for
sidewall cooling (if any) and air in-leakage from sight ports, furnace cracks, etc.
        The primary reason for introducing  air over the fuel bed is to provide
turbulance to the volatile gases that have  evolved from the  solid refuse fuel bed,
thus  improving the combustion of those gases.  A secondary benefit results to the
extent that lower  underfire air velocities reduce the entrainment of fly ash  and
other solid material in the exhaust stream.
 2.3.2   Temperature
        Maximizing  time,  temperature and  turbulence  is  essential  for good
 combustion. Of these parameters, temperature is the only one that can be directly
 measured and also the only one over which the incinerator operator has reasonably
 direct control.
        The temperature  not  only  varies  with  position  within the combustion
 chamber, but also varies with time at a given position.  Most systems monitor the
 temperature (using a shielded thermocouple)  near the  roof at the exit  of  the
 combustion chamber.  Operating temperatures at this location for most MWC's are

-------
in the range of 1200° to 1850° F.  MWC's with waterwalls tend to operate in the
lower end of this range, whereas refractory wall MWC's operate in the upper end of
the range.
       Furnace geometry may indicate a need to monitor temperature in several
locations.  For example,  in rotary kiln  designs, the combustion chamber divides
itself into  a  drying chamber, an ignition chamber, the  rotary kiln, and  finally a
mixing chamber where the furnace gases undergo final combustion.  Temperatures
in such a case are  often monitored at the exit of the ignition chamber and mixing
chamber. The  common operating ranges for these locations are 1600° to 1800° F
and 1800° to 2000° F, respectively.
       Furnace temperature is controlled by adjusting the combustion air supply.
The details as to how this is done  regarding the split between underfire and overfire
air will  vary  among facilities.  In  general,  high  temperatures  are reduced by
increasing the air supply, i.e., quenching the furnace with cool air.  Conversely, low
temperatures are increased by reducing the combustion air.
        Maintaining an adequate furnace  temperature is absolutely essential to
attaining complete combustion and minimizing air pollutants. "Adequate furnace
temperature", in  effect, is the maximum achievable temperature that is consistent
with the long term  mechanical reliability of the system.

2.3.3   Furnace Pressure and Draft
        MWC's  are always  operated at  a  modest  negative  static   pressure,
approximately  0.2  inches of water.   There are several reasons for this. First,
operation under negative pressure eliminates emission of toxic combustion products
through  site  ports, furnace cracks and so forth  into the workplace and general
environment.  All gas leakage related directly to the furnace leaks into, rather than
out of, the furnace. Second, positive pressure exacerbates a phenomenon called
torching. Torching occurs when the flame propagates from the fuel bed downward
through  the grates.  Localized  conditions can cause occasional torching in any
furnace, but operating under negative pressure minimizes the phenomenon and the
resulting problems with heat damage to the grate system and the underfire air
system.
        In order  to remove the gaseous combustion products  from  the combustion
chamber, it is necessary to establish the proper  furnace  draft, which is done by
balancing the airflows  among the over and underfire air systems and the induced
draft (ID) fan.  The ID fan is usually located donwstream of the air pollution control
                                    10

-------
system.   (Some  MWC  systems utilizing wet scrubbers have located the  ID fan
upstream of the scrubbers to minimize  fan  corrosion.) If the furnace  generates
more combustion gases than the fan can handle the furnace pressure goes positive
and  puffing  of combustion products  from the  furnace enclosure can usually  be
observed.  This  condition often  is caused by an  excessive burn rate which can
originate from  an excessive  feed rate  or feed that  has  an  unusually  high
combustible content.

2.3.4   Feed Rate
        In continuous, gravity feed systems,  refuse is piled up in the feed chute.
Refuse in the  chute provides an  air-tight seal while the furnace is in operation.
Refuse is drawn from the chute  by  the action of the grates, either rocking or
reciprocating grates. Thus, the feed rate is governed by the grate speed.  Traveling
grate systems  do not  employ a gravity feed  arrangement.  With traveling  grates,
the feed chute dumps on to an  auger that moves the refuse to the traveling grate.
The  refuse  is  transferred from the auger  to the traveling grate by an air  blast
distributor.  The feed rate depends on the speed of the auger. Augers and traveling
grates are used in systems utilizing RDF.
        Improper feed rates create a variety of problems.  If the feed rate is too
low, insufficient energy is available to maintain an adequate furnace temperature.
If the feed rate  is  too high the furnace  temperature can reach unacceptably high
levels.   In  either  case,  the  bottom  ash will contain a  significant quantity  of
unburned material.
        The feed rate also affects the depth  of the fuel bed on the grates.  A fuel
bed  that  is  too thick will not allow proper penetration of underfire air.   On the
other hand, a fuel bed that is  too thin allows underfire air to penetrate the fuel bed
with  velocities that are sufficient to entrain significant quantities of ash  (and
unburned material)  thus creating opacity and downwind fallout problems.  Excessive
furnace temperatures also cause premature wear and structural problems (buckling,
warping, etc.)  of the furnace refractory  and  grate system  components, as well as
slagging.  Slagging is discussed in Section  2.3.7.

2.3.5   Dealing With Fuel Problems
        This  section discusses procedures for controlling the combustion process
when especially  low or  high  combustible fuel, respectively, is introduced to the
MWC.  Despite the efforts of the crane operator, it becomes necessary from  time
to time to burn fuel that has less than optimal characteristics.
                                    11

-------
       All the  facilities visited  during  this study utilize automatic  combustion
control (trim) systems that employ the furnace exit temperature as the operating
parameter that  controls the combustion  process.  A signal from the temperature
sensor is processed by  a central controller that operates  the dampers on the
overfire and underfire air fans. The details of how such a system operates will vary
from  plant to plant,  but basicaUy, each system responds  to high temperature
excursions by  increasing  the volume  of combustion air in  the  furnace.   The
additional air cools  the furnace.   Conversely,  the  response to low  furnace
temperature   excursions  is  to  decrease  the  volume of  combustion  air in the
furnace.  The MWC visited that utilizes  heat recovery also automatically controls
the feed rate based on steam demand.  There is a trend for newer MWC's with heat
recovery  to  employ oxygen  trim  systems similar to  those  utilized in coal-fired
utility boilers.
        MWC's trim systems will  compensate for moderate swings in fuel quality.
However, if  especiaUy poor quality fuel is fed to the  furnace, it may be necessary
for the  furnace operator to operate the furnace manually  until the combustion
process  has  stabilized.   Whether  manual operation  is required or  not, the
adjustments  to  poor fuel quality  depicted below are  appropriate responses to the
fuel conditions described.

        2.3.5.1    Wet Refuse and Low Combustibles
        Burning wet refuse or low combustibles requires running a shallow fuel bed
on the ignition grates and maintaining high underfire air pressures.  This is often
accompanied by a slight reduction in  the feed  rate in the case of  wet fuel. The
wet, or low  combustible content, of the refuse will effect a reduction in furnace
temperature; thus, increasing the underfire air may  appear contradictory to the
 discussion in  Section  2.3.2 where it  was stated  that  a  decrease  in furnace
 temperature should be  countered  with a decrease in  combustion air.  A  couple of
 points need to be noted.
        First, especially with wet fuel,  the refuse simply will not burn until it is
 dry.   Running a shallow fuel bed with high underfire air ensures good contact and
 mixing of the fuel with the air, enhancing the drying process. Second,  cutting back
 on the feed rate reduces the amount of heat consumed per unit time utilized to
 evaporate the moisture. If the low combustible fuel is not wet, the forge effect of
 the higher underfire  air rate assists in  oxidizing  such refuse.  However, it is not
 usually desirable to cut back on the feed rate as long as the dry refuse is reasonably
 porous.
                                     12

-------
        Automated combustion control systems that control both the underfire and
overfire air will throttle back on the overfire air as the underfire air is increased,
thus minimizing the cooling  effect of the increased underfire air.

        2.3.5.2    Dry Refuse and High Combustibles
        To burn dry, high combustible content refuse, it is necessary to maintain a
high fuel bed on the grates and set a low underfire airflow while maintaining a high
overfire airflow and reducing the feed rate slightly.  Highly combustible refuse
creates excess turbulence in the furnace.  If this  turbulance (which is normally
good) is exacerbated by  a high underfire airflow and a shallow fuel bed, significant
quantities of bottom ash and unburned  waste will be carried out of the furnace and
ultimately  emitted  to  the atmosphere.   The  potential  for   excess  furnace
temperatures is eliminated by maintaining high overfire airflow and reducing the
feed rate slightly.

        2.3.5.3    Undesirable Waste
        The best  solution for dealing  with combustion  problems resulting from
undesirable waste is  to  prevent  such waste from ever reaching the  furnace.   The
kind of waste that is usually amenable to this  solution is certain kinds of highly
combustible waste such as rubber tires, mattresses and certain types of commercial
and industrial waste.
        Some municipalities do  not allow neighborhood trash collection crews to
pick-up tires.  Such  prohibitions often include  items that can cause mechanical
damage to the MWC  such as large appliances, axles, engine blocks, etc. Other
municipalities pick-up all trash  and leave it up to  the crane operator to pick out
undesirable refuse.

2.3.6    Grates
        The purpose of the grate system is to transport the refuse and residual ash
through the combustion  chamber. The grates are designed to allow the passage of
underfire air from  underneath  the grates  up  through the  grates and fuel  bed.
Certain grate designs (rocking grates and reciprocating grates) provide a significant
amount of agitation of  the refuse as it  tumbles  from one grate section to the
next. Such agitation helps promote  complete combustion. Traveling grates provide
less  agitation and are used primarily for RDF.  In  the United States, most stoker
fired MWC's utilize reciprocating, rocking, or traveling grates.
                                    13

-------
        Regardless of which of the three grate types are employed, the grates can
be qualitatively divided into three sections.  The first section is referred to as the
drying or  charging grate.  Most of the drying in this section is a result of radiant
heat from the furnace arches and walls.  The second section is referred to as the
ignition grates.  The  refuse  is ignited in this section and a  significant  amount
(approximately 50 percent) of the burning takes place in this section. The third and
final  section is referred  to as the burnout or combustion grates.  Combustion is
completed  in  this section.    The   remaining  residue  (bottom  ash and non-
combustibles) falls off the end of the burnout grate, is quenched, and conveyed out
for final disposal.  (Ash handling  is discussed in Section  2.4.)  If a rotary kiln is
employed, it normally is preceded by drying and ignition grates of the reciprocating
type.  The kiln serves the  same purpose as the burnout grates.
        From an  operating  point  of  view,  the  main  concern about  grates  is
preventing  mechanical and thermal damage to  the grate system.   Mechanical
problems  include damage caused by allowing excessively heavy items to get into
the furnace.  Engine blocks, automobile transmissions and even refrigerators have
been known to find their way  into MWC's. As such heavy items tumble through the
furnace they can bend, break, or otherwise severely damage the grates, sometimes
requiring the plant to shutdown. Smaller material such as  chains, wire, small pipes,
etc. fall through the grate system  and if not cleaned out periodically  can jam the
grates, requiring a shutdown.
        Excess  heat can  warp or buckle  the  grates  and the  support  system
underneath  the grates.  This problem  can be  caused by high furnace temperatures
or, probably more commonly, by excessive torching.

2.3.7    Slagging
        Excessive furnace temperatures can cause slagging. Slagging occurs when
temperatures in the furnace  (or some portion thereof) are sufficient to fuse the
ash.  This causes the ash  to run and to evaporate, but it will solidify or condense at
the first opportunity. Different types  of refuse produce different types of slag that
have  different melting and running temperatures.  Slagging creates havoc with the
grate systems by filling underfire air slots, thus creating poor air distribution. Slag
can also  interfere with  the  movement of the grates themselves.  If the grates
become  completely frozen  due  to   solidified  slag,  an  immediate  shutdown  is
required.
                                    14

-------
             Slag formation on the furnace walls win cause premature deterioration of
     the  refractory.  Many MWC  operators  monitor for slag formation by regularly
     observing the  appearance of the furnace refractory in  the lower portion  of  the
     furnace.  Fused ash sticking to the furnace refractory gives the walls a  wet or
     glassy appearance.

2.4  ASH HANDLING
             Many MWC operators report that the ash handling equipment creates more
     O & M problems on a day in - day  out basis than any other system in the facility.
     As discussed  below,  some  of  these  O  &  M  problems  result  in air pollution
     compliance difficulties.

     2.4.1   Ash and Residue Composition
             If the residue produced by MWCs contained only bottom ash and fly ash  the
     problem of ash handling would be much simpler than it is. Unfortunately, however,
     the  residue contains a substantial fraction of non-combustible material, including
     glass fragments, metal material (cans,  wire, axles, springs, sheet metal, motors,
     etc.), clinkers, rocks, and unburned  organic material. This  collection of variously
     sized material is abrasive and difficult to  handle.  Ash can  be  acidic or basic,
     depending on its source.

     2.4.2   Ash Collection Points
             The number and location of ash collection points will vary somewhat  among
     MWC's,  even  if  the  facilities are  of rather similar design.   The following five
     collection points are common to  many facilities in  the United States.

     1.      Grates - Siftings are the ash and residue  that fall through and are collected
             underneath the grates.
     2.      Quench Pit - The quench pit collects the bottom ash that faUs off the end
             of the  combustion grate. The quench pit contains water so that the ash is
             cooled for further handling.
     3.      Mixing Chamber - The mixing chamber holds the volatile furnace gases at
             the  proper  temperature long  enough to  allow  complete  combustion   to
             occur.   Heavy flyash falls  out of the gas stream at this  point. In some
             designs, this fallout is collected in the quench pit, i.e., the  mixing chamber
             is directly   above the quench pit.   If the  mixing chamber  is slightly
             downstream  of the  quench  pit,  it is necessary  to collect  the  ash  at the
             bottom of the mixing chamber and evacuate the ash through a rotary  air
             lock valve.
                                        15

-------
4.      Cooling Tower - Cooling towers  lower the temperature of the furnace
        exhaust gases  via evaporative cooling prior to the gases entry into the air
        pollution device.  The cooling towers often function as ash collection points
        because the tower  vessel also acts as  a gravity collector.  To  a limited
        extent, they also perform as prescrubbers.  Sometimes cooling towers are
        referred  to as quench  towers or chambers, and if they precede  a wet
        scrubber they  may be referred to as a presaturator, if saturation of the gas
        stream is their function.  Most MWC's with heat recovery do not utilize
        cooling towers.
5.      Air Pollution  Control  Equipment  - Electrostatic  precipitators (ESP's),
        baghouses and  scrubbers^ of course, collect the fly ash produced by MWC's.

        Rectangular batch furnaces  typically  have ash collection  and storage
hoppers directly underneath the grates.  Periodically, ash and residue are removed,
quenched, and accumulated in a residue hopper. Discharge from the residue hopper
normally is done through a watertight gate valve into  trucks  or other containers
and shipped to a disposal area, usually a landfill.

2.4.3    Ash Conveying Systems and O & M Considerations
        The means of  moving the ash and residue produced at  MWC facilities can
be broadly classified  as either manual or continuous, with some plants using a
combination of these  techniques.  This section focuses on  ash handling problems
that will adversely effect atmospheric emissions.
        Quench Pit - Continuous ash removal operations employ a variety of means
to move ash through  the plant.  The bottom ash moving off the grates is usually
dumped into a quench pit.  The water in the pit  actually provides an airtight seal
for the furnace.  If the water level in the pit is not maintained at the proper level,
the seal is broken and it becomes impossible to  maintain proper furnace pressure
and draft.  A drag conveyor, submerged at its lower end in the quench pit, pulls ash
from the pit to a holding hopper or in some cases directly into  trucks for transport
to a landfill.  Since the drag conveyor is inclined, it allows some of the entrained
quench water to drain back to the quench pit.  Often, these drag conveyors become
hung  up with the odd  sized material that accumulates in the quench pit. If these
hang-ups occur above  the water line they can usually (but not always) be cleared in
a short time without interrupting operation of the furnace.  If they occur below the
waterline,  it is usually  necessary to drain the quench pit to  gain  access  to  the
problem source.   Draining the pit will break the airseal and result  in  higher
emissions until the problem is corrected.
                                    16

-------
        Grates - Siftings are removed from under the grates by vibrating or screw
type conveyors, or by water sluices. The collected material is sent to the quench
pit.  Most facilities will observe the siftings  conveyors area on an hourly basis to
ensure that excessive amounts of material are not catching on beams and the grate
drive system or jamming the conveyors. Sometimes such problems can be resolved
by opening inspection doors and knocking the material off with  long poles.  This
draws a lot of cold  air into the  furnace  and thus  usually creates an  opacity
problem.  If solving the problem requires shutting down the furnace, even  greater
air emissions result.
        ESP Hoppers -  Fly ash from ESP or baghouse hoppers  may be  dumped
through an airlock valve directly to a sluicing system, or it  may be moved by a
screw conveyor system prior to being dropped into a quench pit.  Condensation in
the hopper area can cause bridging of the ash, and can even cause the ash to create
clinkers that adhere to  the hopper walls.  When the  ash backs up into  the ESP's
electrical field it  causes excessive sparking and eventually the ESP will  short out.
Most commonly, bridging and clinkering are caused by air in-leakage (ESP's are
operated under  negative pressure)  in or near the hopper  area.  Cool ambient air
leaking into the hopper  condenses  water vapor and acid gases.   Another cause is
excessive cooling  at the cooling tower, or cooling tower spray  nozzles  that have
deteriorated to the extent that they produce  large droplets that cannot completely
evaporate prior to entering the ESP. Yet another  potential cause is leaking boiler
tubes in MWC's equipped with heat recovery.
        Sometimes the subsequent bridging can be rodded out  while the MWC  is in
operation.   If this fails, it  is necessary to shutdown the facilitly to gain internal
access to the hoppers.
        Some operators report that bridging is sometimes caused by sticky material
such  as plastic film reaching the hoppers without being burned.  This may be
indicative of excessive airflows, especially underfire air, in the furnace.
        Prevention of hopper problems entails aggressive maintenance  procedures.
The  ESP's,  hoppers  and  air lock valves need to be inspected for leaks at every
opportunity.  The air lock valves should be externally inspected and lubricated
weekly when the MWC is in  service.
       From a  design  perspective, ash  bridging  and  flow problems  can be
minimized through the installation  of vibrators on the  hoppers.  Condensation can
be minimized by insulating the  hoppers or installing hopper heaters.
                                    17

-------
             Mixing Chamber - The rotary air lock valve should be inspected for leakage
     internally at each opportunity and inspected and lubricated weekly when the MWC
     is in operation.
             Summary - Basically,  the important components of the ash handling system
     should be monitored frequently to ensure that residue is not backing up and thus
     creating a  potential shutdown  condition.   These observations include checking
     water levels,  flow  rates, pumps, motors  for rotary valves, and free movement of
     conveyors and residue.
             The details of specific systems will vary from the description  above.  For
     example, facilities  that recycle  process water  may send sluice residue directly to a
     clarifier, rather than to the quench pit.  Also, recycle systems sometimes control
     pH of the process water by use  of chemical additives.  Regardless of the details of
     the  methods  and equipment  employed  at a  given  facility, clean  and efficient
     operation of the MWC is tied directly to  the reliable operation of the ash handling
     system.

2.5   AIR POLLUTION CONTROL EQUIPMENT
             This section  discusses  the  control  devices  encountered during  the  site
     visits: electrostatic precipitators (ESP's) and high  energy wet  scrubbers.   The
     discussion  is limited  to the specific concerns brought to our  attention by MWC
     operators.   Regardless  of the type of control device, a common concern expressed
     was corrosion control in the  control equipment and ductwork.  (Broader,  in-depth
     discussions of control equipment design, O & M, and inspection, recordkeeping, and
     performance evaluation are discussed in the  literature.  For examples,  refer to
     References 2, 3, 4,  5, 6, 7 and 8.)

     2.5.1   Electrostatic Precipitators
             ESP's are the  most common particulate matter  control device applied to
     MWC's today.  This section will briefly discuss corrosion problems  and useful
     monitoring and recordkeeping techniques for ESP's.
                                         18

-------
        2.5.1.1   Corrosion: Causes and Prevention
        A successful corrosion prevention program must first recognize that, as a
result of the combustion reaction,  MWC exhaust gases contain a significant water
content.  Also, water is added  to  the exhaust gases at many facilities either  by
design or  accident.  The acid content (mostly HC1) of the furnace exhaust gases
further complicate the problem.
        The effect  of excess moisture on  the ESP's ash hoppers was  discussed in
Section 2.4. The corrosion problem discussed here manifests itself primarily as ESP
plate corrision. Within a very few  years holes of several feet in area can develop.
This requires municipal authorities  to decide whether to engage  in costly repairs or
limp along with reduced plate area.  Indeed, some  MWC operators have expressed
the opinion that a  major ESP  rebuild is required within seven years for  units in
MWC service.  As operating experience  has accumulated over the years, however,
it becomes apparent  that  corrosion in MWC ESP's can be controlled to levels
similar  to those in other industries  where rebuilding of the unit is not necessary for
at least 15 to 20 years.
        Cooling Tower - The  purpose  of  the  cooling tower  is  to lower  the
temperature of the exhaust  gases to a point where the ESP will  not suffer  thermal
damage.  (Plate warping and buckling are the most common heat related problem.
In more severe temperature excursions,  the ESP's structural members can also  be
damaged.) The cooling is accomplished by evaporating a fine mist introduced into
the gas stream by  water  sprays  in the  cooling tower.   If too  much water is
introduced into the gas stream, all of the  water cannot evaporate before entering
the ESP, thus creating a potential corrosion problem.
        If the spray nozzles in the cooling tower have  eroded  substantially, they
may produce a course spray (large droplets) that cannot completely evaporate prior
to entering the ESP.  Again, moisture in the ESP chamber  establishes a corrosion
hazard.
      •  A thermocouple is placed near the entrance to an ESP.  The signal is sent
to the control panel so that the MWC operator can monitor the  inlet temperature.
The signal is also sent to a controller that governs the quantity of water supplied to
the cooling tower.  It  is imperative that the operator  monitor this system so that
he can take appropriate action if the performance of the cooling  tower  deteriorates
for any reason. The cooling tower spray nozzles and the thermocouple should  be
inspected  at every scheduled outage for MWC's that operate for at least a month
continuously.
                                   19

-------
       Air In-Leakage - Since ESP's operate under negative pressure any leakage
through cracks or pinholes on the enclosure will allow air to move from the outside
to the inside.  Cool  air coming into the  unit  is likely to cause condensation of
moisture and acid gases.  If plate  corrosion is confined to a limited area, air in-
leakage should be suspected since airflow  in the ESP is not turbulent; air streams
from holes and cracks tend to  maintain their integrity, do not diffuse very  much,
and thus tend to attack rather small areas.
       The insidious  characteristic of air  in-leakage, however, is that it tends to
feed on itself.  If an air  stream from a leak flows  in a manner  that causes it to
contact the shell of the ESP, the ensuing condensation  will cause  corrosion at that
location, ultimately creating another  route for leakage. Another common source of
leaks are deteriorating seals on access doors.
       The ESP should be thoroughly  inspected for leaks at least  on a quarterly
basis if the operating schedule allows it.  It is  important to correct any problems
found during the inspection before  the system is brought back on-line. Otherwise,
the situation will likely be much more severe at the next shutdown.
        As unlikely as it may seem, it is advisable to  inspect a new ESP prior to the
initial  startup to ensure that the field erection crews  have welded  all appropriate
seams  in  the  ESP  shell and hopper  areas.  MWC and other ESP operators have
learned this technique through the school of hard knocks.
        Shutdown - Every time an  MWC is shutdown,  the hot, water vapor  laden
exhaust gases in the ductwork and ESP must cool to below the dewpoint.  In- effect,
the ducts and air pollution control  device take an acid bath every time the system
is shutdown.  Obviously, the corrosion this process inevitably causes is more severe
if shutdowns are frequent. Facilities that  regularly shutdown on a weekly schedule
probably have the most severe problem.
        Two steps can be taken to alleviate condensation during shutdowns.  The
first is to minimize  the  number of  unscheduled shutdowns by  implementing an
aggressive O & M plan, particularly on those parts of the system  whose failure can
cause a shutdown situation, e.g., the ash handling equipment, the  grate drives, and
the air pollution control system itself.
        Second, if the normal operating schedule calls  for a weekly shutdown, it is
then feasible  to install an ESP heater. An ESP  heater is basically an oil  or gas
burner installed in the duct on  the  inlet side of the ESP.  The heater is fired up for
short term  (weekend) shutdowns, preventing the  exhaust gases  in the ESP from
condensing during the shutdown period.
                                    20

-------
        Performance Monitoring -  Virtually all MWC operators who employ ESP's
do some performance monitoring.  The commonly monitored electrical data include
primary current, primary voltage, secondary current, secondary voltage, and spark
rate.  Many older units do not monitor secondary voltage and some of the newer
systems do not include a spark rate guage.  Regardless of the details as to precisely
what data is available for a given unit, most operators record the electrical data on
an  hourly  or  bi-hourly  schedule.  Table  2-2  is  a typical log for recording the
secondary electrical data.
        The ESP  electrical data is  a powerful  tool  for evaluating the overall
performance of the unit and in diagnosing performance problems.  The  reader is
referred to the EPA publication in Reference 2 for an excellent discussion of the
utility of these parameters.
        Another  useful parameter for  evaluating  ESP performance  is  opacity.
Federal New  Source Performance Standards (NSPS)  do  not require continuous
monitoring of  opacity  at MWC's.  Indeed, some MWC's normally operate  with
condensation  occurring  downstream of  the control  device, making continuous
opacity monitoring  unfeasible.  Yet many other  units  do not create such steam
plumes. Thus,  many states have incorporated continuous opacity monitoring permit
requirements on a case by case basis.   Where opacity data is available  it can be
very useful in evaluating performance and diagnosing problems. Again, Reference
2 provides an excellent  discussion on how to utilize opacity data as a diagnostic
tool.

2.5.2    Wet Scrubbers
        High efficiency  scrubbers for  MWC's include  variable  and fixed throat
venturi scrubbers and  flooded disc type  venturi  scrubbers.  One of the major
problems with  wet  scrubber performance  as applied to MWC's is  control of the
condensible,  partially  oxidized  organic  matter  which forms when  incinerator
operating temperatures are too low. The organic material condenses while passing
through the  scrubber,  creating  an aerosol with  a particle size  too  small for
effective impaction in the scrubber.  Few  scrubbers have been able to achieve the
NSPS emission limits because of this problem (Reference 3).
        Corrosion is a  major concern  in  wet scrubber operations, as it is  with
ESP's.   For scrubbers,  the solution is  centered  around control of the  scrubber's
liquor quality.  The paragraphs below briefly discuss the impact of liquor quality on
system performance.
                                   21

-------
                                  TABLE 2-2
                      ELECTROSTATIC PRECIPITATOR LOG
                             Incinerator No.
                                 Date
                  Field No. 1
                               Field No. 2
           Secondary
            Voltage
             (KV)
Secondary
 Current
  (Ma)
  Spark   Secondary
  Rate    Voltage
(per min.)    (KV)
Secondary
 Current
  (Ma)
  Spark
  Rate
(per min.)
   1 am
   2 am
   3 am
   4 am
   5 am
   6 am
   7 am
   8 am
   9 am
  10 am
  11 am
   Noon
   1 pm
   2 pm
   3 pm
   4 pm
   5 pm
   6 pm
   7 pm
   8 pm
   9 pm
  10 pm
  11 pm
Midnight
                                      22

-------
        Liquor Quality,  Corrosion, and Performance - From both  a performance  and
maintenance perspective, liquor quality is  critical to the scrubber's reliable operation.
The critical parameters are pH, and suspended and dissolved solids.  The rigor with which
a source operator monitors these parameters may not need to be as severe for a scrubber
utilizing a once through system, but most operations recycle a significant fraction of the
spent liquor; some recycle all  of  it with fresh water  introduced  only to make up for
evaporative losses.  The concern for liquor quality applies to both the liquor supplied to
any cooling tower or presaturator and to liquor supplied to the scrubber itself.
        In municipal incinerators,  scrubber liquor tends to become acidic  due to  the
absorption of HCl'and organic acids in the  liquor. Acidic conditions can quickly lead to
serious   corrosion  problems  everyplace  downstream  of  where  the  acid  liquor is
introduced.  Thus, the spray chamber, ducts, scrubber, entrainment separator, ID fan and
stack can suffer  corrosion  problems.  In areas of  the system  that  are under  positive
pressure, (i.e., the  scrubber  is  preceded by  the ID  fan)  the  holes and cracks from
corrosion in  the system  will  allow  untreated  combustion  gases  to  escape  to  the
environment.  For those  portions of the system under negative pressure, holes and cracks
in the  system can  substantially  reduce the  ID fan's  ability to control draft in  the
furnace. In addition, in-leakage of ambient  air causes localized condensation of moisture
and acid gases, the resulting liquid exacerbating  the corrosion problem.  Thus, you have
something similar to a monster feeding on itself.
        Liquor  pH   and water  make-up   rate can  be  monitored  and  controlled
automatically  at  the clarifier just before the  recycled liquor is  reintrodueed to  the
system.  The pH of the liquor should be maintained above the levels at which carbon steel
is attacked.  A pH of 6  or greater is usually satisfactory.  Chemicals utilized to reduce
pH are usually sodium hydroxide, ammonia, or lime.
        The dissolved and suspended solids content of recycled liquor varies from source
to source, but can be as high as 15% by weight.  Since a major portion  of the liquor
injected into the quench tower and/or presaturator  evaporates, some of the solids
entering as suspended and dissolved solids in the liquor are released as particulate matter
(Reference  11).  High solids content  also is  often responsible for  excessive erosion of
pipes and pumps,  and pluggage of spray nozzles.
        Samples  of  the  liquor  should  be  analyzed  periodically, even if  this requires
sending the  samples  to an outside laboratory.  High solids content in the liquor can be
deviated by several  methods including more frequent  blowdown of the scrubber liquid
system and a lower recycle rate.
                                         23

-------
        Proper control of other scrubber variables such as pressure drop and liquid-to-gas
ratio  are also critical  to  maximizing scrubber performance.  For  a more detailed
discussion of these parameters, the reader is referred to Reference 3.

2.6  FANS, DUCTS, AND STACKS
             As  auxilliary  systems the importance  of  fans, ducts and  stacks  to  the
     overall operation and performance of  an MWC is  easy to underestimate.  This
     section briefly describes some of the salient O & M concerns for these auxilliary
     systems.

     2.6.1   Fans
             Operation  and  maintenance  considerations  for  fans depend  on several
     factors.  These factors include the type of service (and locations) the fan is in, fan
     size, fan type, and whether the fan is a  direct drive or belt driven unit.  Regardless
     of any general "rules of  thumb" provided here or elsewhere, source operators should
     follow  the  specific  fan  manufacturer's  recommendations  for  the nature  and
     frequency of fan maintenance procedures as minimum requirements for reliable fan
     performance.
             The fans of principle concern at municipal incinerators are the combustion
     air  fans  (underfire, overfire, and if applicable, sidewall  cooling  fans) and  the
     induced draft  (ID)  fan.   Generally, the  combustion air fans are constant speed
     devices with  airflow  controlled by dampers.   The ID  fans may also be constant
     speed units, but variable speed units are probably more common.
             Most ID  fans are located between the air pollution  control system and the
     stack.  If the ID  fan precedes the air pollution control equipment, problems related
     to abrasion, erosion,  corrosion, and  build-up of solids on the fan  will be more
     severe.  The only "solution" to this is frequent inspection and cleaning.
             ID  fans serving  scrubber systems operate under substantially more severe
     conditions than those  serving dry systems. This is due to the humid conditions that
     enhance the possibility of condensation of acids on the fan and fan housing. Even if
     the ID  fan preceeds the scrubber, the fan is usually placed after the cooling tower,
     since the cost of a fan that can handle the substantially higher  gas volumes and
     temperatures that occur upstream of the cooling tower would be prohibitive.
             Regardless  of its location, the ID fan is usually equipped with water cooled
     bearings  and a vibration sensor that will sound an alarm in the plant and illuminate
     emergency lights in the  control room  if vibration becomes excessive.  In some
                                         24

-------
systems,  the  sensor controls  will automatically shutdown the fan and  open an
emergency stack in preparation for an emergency shutdown.  Aside from  the fact
that the  furnace  cannot operate properly  without  the ID fan,  everyone  working
around  industrial  fans should be aware that vibration is a preliminary indication
that the fan might shake apart at anytime, possibly throwing fan parts considerable
distances at substantial risk to life and property.
        Fan current  for  all fans is often  monitored in the control room.  At a
minimum, this information  tells  the incinerator operator  whether fans  are
operating or not and can provide rough estimates of airflows.  These meters may
also include audible alarms or emergency lights that indicate no current. If the ID
fan  loses current,  many  systems  will,  again,  automatically begin shutdown
procedures.                                                                 \
        Table 2-3  illustrates a typical fan inspection and preventative  maintenance
schedule.   Operating experience  or the  manufacturer's recommendations may
dictate a more frequent inspection and maintenance  than indicated here.

2.6.2   Duets
        Air ducts  are a passive part of  the incinerator  system.  Nevertheless,
proper  maintenance  of  the ducts  is  critical  to  the  overall performance of  the
system.  Two problem areas are addressed in this section, leakage and deposition.
        Leakage - Leaks in air ducts  result from any of several causes, including
corrosion, abrasion/erosion, flex failure and failure  of expansion joints.  Leaks can
occur as holes, cracks, and fissures in the duct system. The causes and effects of
such leakage will  vary somewhat, depending on whether  they occur in ducts that
are under positive or negative pressure.                                       :
        Positive pressure ducts occur  downstream of the ID fan. In most systems,
this  is only a small fraction of the total duct system.  A few incineration systems
however, locate  the ID  fan  upstream  of  the air  pollution control  device, thus
creating substantial duct lengths between the fan and control device and continuing
on to the stack.  In a positive pressure  environment, leaks occur from the duct to
the ambient atmosphere. Between the fan and the control device,  this means that
untreated furnace gases are exhausted directly to the atmosphere.
        Air in-leakage occurs in negative pressure ducts.  Negative pressure ducts
occur upstream of the ID fan. In most systems, the ID fan is the last unit operation
prior to the stack, so most of the ductwork,  in fact,  most of the entire system,
including the furnace, the air pollution control device and cooling  towers, operate
under negative pressure.
                                   25

-------
                                      TABLE 2-3
       TYPICAL FAN INSPECTION AND PREVENTATIVE MAINTENANCE SCHEDULE
INSPECTION/MAINTENANCE ITEM
DAILY
WEEKLY
MONTHLY
 SEMI-
ANNUAL
Vibration Check
Oil Level
Oil Color
Oil Temperature
Lubricate
Bearings Check
    Noise
    Leaks
    Cracks
    Loose Fittings
    Inspection (Internal)
    (Clearances, Wear,
    Pitting, Scaring)
Clean and Inspect Blades and
    Internal Housing
Fan Belt
    Noise Check
    Belt Tension and  Wear
  X
                  X
                  X
                  X
                  X
  X
                                X
                                X
                                X
                                             X
                                            XJ
  X
   Whenever the fan is out of service
LFans in scrubber service may require cleaning on a monthly or quarterly schedule.
                                      26

-------
        Air in-leakage creates several problems.  The first is that it tends to feed
on  itself.   Leakage of  cold  ambient air  into  the  ducts  encourages  localized
condensation of acidic vapors that tends to create more holes and cracks. As this
situation propagates, the ID fan attempts to pull more and more air.  It becomes
difficult for the furnace  operator  to maintain the  proper  furnace pressure and
draft.                                                                       I
        Ducts should be inspected externally for  leaks  on a monthly basis,  more
often if the facility has a history of leaky ductwork.  Small leaks are  difficult to
locate by  an external inspection.  Internal  inspections performed during daylight
hours have the  advantage of highlighting small holes and cracks due  to contrast
between the dark duct interior  and bright daylight.  Internal inspections  should be
performed on a quarterly schedule.
        Particle  Deposition  - Dust  build-up  is an  inevitable difficulty in  many
horizontal duct sections  upstream of  the air pollution  control device due to the
impractical economics of maintaining air velocities that are sufficient to  carry the
very largest particles in the gas stream.   Consequently, many  designers provide
conveniently located clean out ports so that  the build-up can be easily  removed.
Excessive  build-up  in MWC  systems  result from  oversized ductwork,  or,  more
commonly, excessive underfire air  and/or  furnace  draft   that  tends  to   carry
significant quantities of unburned  material out of the  furnace and into the air
handling system.
        The consequences of excessive dust build-up are several.  First,  by reducing
the  cross-sectioned  area of  the duct,  build-up  increases air  velocity.    This
consequence alone  can  deteriorate the performance  of the air pollution control
device.  Second, by moving the entire air stream over to one portion of  the  duct,
maldistribution problems in ESP's are likely to occur, despite turning vanes and gas
distribution plates, which were never designed to correct for excessive dust build-
up.  Finally, duct systems are not usually designed as load bearing structures.  The
build-up of excessive quantities of material in air ducts can  (and sometimes  does)
lead to collapse of the duct support system or the duct itself. Aside from requiring
an  immediate  shutdown and  rather  expensive repair,  collapse of  ductwork  is
potentiaEy dangerous to plant workers.
                                    27

-------
2.6.3   Stacks
       Stacks  are  another  passive  system  from  an  operational procedures
perspective.  Air pollution agencies are concerned about only a few specific aspects
of the initial design.  First, of course, is the stack height. The function of stacks is
to lift incinerator effluents to a suitable elevation above ground level so that the
air pollutants remaining in the effluent can be adequately dispersed and diluted
prior  to being returned to  ground level.  Second, if the stack is to be the sight of
stack sampling ports and/or pollutant monitoring  instruments (transmissometers,
S(>2 gas analyzers, carbon  dioxide monitors, etc.) then the design must provide for
safe access to the ports and instruments, including appropriate scaffolding.
       Twenty or more years  ago it  was common  for  municipal incinerators to
operate without air pollution control equipment. In those days, MWC's operated by
natural draft, rather than  with ID fans.  The installation of  air pollution control
equipment required  cooling the incinerator  effluent to provide temperatures in  a
range that would not damage scrubbers, baghouses, or ESP's.  A second important
aspect of  the gas cooling was the reduction in gas volume  that followed.  The
cooler, slower gas stream now  required an induced draft fan so that the incinerator
operator could maintain proper furnace draft and pressure.  The complications this
created for stack  design relates to the increased potential for condensation of acid
containing vapors  on the  stack  liner.   Most  incinerator operators schedule  a
thorough  stack inspection  on a yearly basis.   The inspection addresses  both the
structural integrity of the stack and  chemical attack  of the liner.   Often these
inspections are conducted by outside firms who specialize in stack maintenance.
       Observation of the physical stack can sometimes assist in identifying  or
diagnosing problems elsewhere in the incinerator  system.   Stacks in  scrubber
service, for example, sometimes develop a mud lip, that is, the stack exit becomes
covered   with  wet  particulate  matter.   This  is  generally the  result  of  a
malfunctioning entrainment separator.
       Some operators have been known to cut "windows" in the stack near the
base. The purpose for doing this has never been clear.  Aspirating ambient air into
the stack  lowers  the stack gas temperature,  thus increasing problems related  to
condensation  of  acids in  the  stack.   In addition,  the lower gas  temperatures
decrease plume bouyancy,  thus decreasing the effective stack height and resulting
dispersion at ground level. Finally, in most jurisdictions, dilution of the stack gases
in this manner is considered a circumvention of the opacity emissions limitation.
                                    28

-------
2.7  STARTUP AND SHUTDOWN                                                 ;
             Startup and  shutdown  are routine activities at MWC's  that  can cause
     periods of excess emissions. The excess emission problem is much more severe for
     startup than it is for shutdown. This section describes operating procedures that
     can minimize excess emissions during startup and shutdown.

     2.7.1   Startup
             It is  common for  newer  MWC's to use  auxiliary burners as a means of
     achieving a prescribed minimum furnace temperature before waste  is charged to
     the system.  However, for MWC's which do  not utilize auxiliary burners, the typical
     startup is a  procedure  in  which  the refuse loading and the combustion air are
     gradually  increased,  and  during  which  time  the  furnace  temperatures  are
     intentionally  controlled below  the levels required for optimum combustion.  The
     initial charging rate  normally varies from  one-third to one-half of the design rate
     and is increased stepwise over the duration  of the startup period.
             During startup, particulate matter  and opacity emissions can  be extremely
     high.  This is  due to the low furnace temperatures that exist for the first two to
     four  hours.   The  result is incomplete combustion  and the generation  of large
     quantities of soot.   The reason for the long startup interval is  the necessity to
     slowly heat the MWC refractory material and downstream components and thereby
     prevent damage from rapid expansion.  Rapid temperature increases can result in
     spalling and premature  failure of  the refractory and similar heat stress problems
     for the grate system.  For units equipped  with boilers, all of the pressure parts of
     the boiler must  be heated  and  gently  expanded  into their  position  at  full
     operation.  Typically, the rate of temperature increase at the boiler outlet is
     limited to 100°F  increase  per  hour as a uniform gradation in temperature, i.e., a
     25°F increase in boiler outlet temperature every 15 minutes.   Since the boiler
     outlet temperature at full load operation  conditions is  approximately 650°F, this
     would result in a startup period of approximately 6 hours.  Excessive opacity will be
     generated for the first 2 to 4 hours. On the more sophisticated MWC systems, it is
     possible to reduce the startup  time somewhat by bypassing the economizer  or ,air
     preheater at  the tail end of the boiler so that the flue gas temperature is increased
     as rapidly as possible to the air pollution control equipment.
             Electrostatic Precipitators - During the initial 4 hours of startup, the flue
     gas products  going   to  an ESP   can  be  wet  and  sticky due to  16w  flue gas
     temperatures.  If the ESP is placed in operation at the time the furnace is ignited,
                                         29

-------
it can be difficult to remove the sticky particles from the plates.  As the flue gas
temperature increases these particles can bake onto the plate. This situation can
reduce the efficiency of the ESP when the MWC is fully charged. This is why many
manufacturers of electrostatic precipitators do not recommend power be turned on
until  the acid dew point temperature has been exceeded at the ESP outlet.  If the
MWC is equipped with ESP  heaters as described in Section 2.5, the heater can be
used to preheat the ESP prior to furnace charging, thus allowing the control device
to be energized much earlier in the startup process.
       Economizers  - In order to exceed dew point temperature in the control
equipment  as quickly as possible,  a bypass of the water around the economizer is
required so that the  economizer will not  reduce the flue gas temperature during
startup.  Most economizers are equipped with bypass piping so that the feedwater
can be bypassed and not permitted to pass through the  economizer.  This can
decrease the time  necessary  to achieve an  adequate flue  gas temperature by as
much as 2 hours.  The problem with this procedure occurs in putting the economizer
into service once the boiler has been brought up to pressure and placed on line.  At
this time the economizer will generate steam when feedwater is introduced into it
and this  steam  must be vented out through the vent  valve  at its top until"the
steambound economizer can begin to circulate water to the boiler without being
injurious to the boiler.  This requires a high degree of operating skill "on the part of
the plant personnel.
       Air Heaters  - Air  heaters  can also be bypassed.   This requires a duct
between the discharge of the forced draft fan  and the windbox area of the stoker.
This prevents the forced draft air from being preheated by the flue gases, therefore
allowing a more rapid  increase of the  flue gas temperature at the  air polltition
control equipment.
       Multiple  Furnace  Facilities -  An effective way  to eliminate opacity
problems  during startup  at  multiple  furnace  facilities  is  to  provide  inter-
connections among the various furnaces and the air pollution control equipment.
With  such a setup, the air pollution control equipment (ESP  or baghouse) associated
with  a boiler coming on line can be preheated to a point above the acid dew  point
temperature  by  a  slip stream(s) from the  flue  gas  of other  fully operating
furnaces.   Thus, baghouses and ESPs can be brought on-line much earlier in the
startup  process.    Maintenance  of damper  seals is very important in  an
interconnected setup so that the potential for air infiltration and out-leakage is
eliminated.
                                   30

-------
       Superheaters  - During  startup  of  a unit,  the superheater section and
reheater sections could be flooded with feedwater if hydrostatic testing has been
conducted. If so, the refuse feed rate into the furnace must be held low enough to
allow the water that was in the superheater (or reheater section) to be converted to
steam and vented  to  the atmosphere.  During  the initial startup  of  the boiler,
thermocouples are normally attached to determine the time it takes to boil away
all  of the water  in each and  every superheater  or reheater element.  The refuse
feed rate must be limited to ensure that the superheater and reheater elements are
heated and cleared slowly so  that saturated steam may pass through them and not
cause overheating of superheater metal. This time/temperature curve is drawn up
by the boiler manufacturer for each and every boiler at initial startup.  Time must
be allowed, especiaUy at the very low firing rates where excessive opacities can be
generated, to clear  these units.   After clearing the  liquid water  from  the
superheater,  heating the superheater metal  is  usually limited to  50°F  per hour.
Thus, clearing liquid water and gentle heating of the  superheater usually requirfes
about ten hours.   After this ten hour period, the stack gas is hot enough to allow
energizing the ESP or bringing the baghouse on-line.
        Short-term Shutdowns - The MWC startup time can also be greatly reduced
if the MWC  is to be shut down  for a short time over a weekend, by converting the
MWC and air pollution equipment into a  "thermos bottle".  This requires (1) the
complete stoppage of the forced draft fan air flowing through the furnace after the
MWC has been brought off line and steam is no longer being generated in the unit
and (2) not allowing any air leakage into the setting and not allowing natural stack
effect of the high temperature flue gases (i.e., gases at 300°F or 400°F should riot
be allowed to pass through the MWC to heat  the tailend economizer, air heater, or
ID fan,  and then pass out the stack).  By making a "thermos bottle", the MWC will
 minimize the loss of heat and fast startup  (3  to 4 hours) can be accomplished.
 When startup is  initiated  the  following Monday  morning, the  internal MWC
 temperature  is  already  up  to  250°F to  300°F instead  of the  ambient  air
 temperature of 70°F to 80°F.
         Most operators do not make the effort to stop the natural draft through the
 MWC,  thereby cooling the furnace  and increasing the length of time  for system
 startup. It  is most desirable to  prevent a MWC system from cooling down if the
 unit is  to be off line for a  short period of one to three  days.  Such cooling  is
 undesirable because it takes  more time to startup a cold furnace.  Also, a furnace
 that is allowed  to   cool  down from operating temperature to  ambient  air
                                    31

-------
     temperatures (e.g. 70°-80°F) requires more maintenance than a unit that cools to
     only 250°-300°F. The greater and more frequent the expansion and contraction of
     the pressure parts and refractories, the more maintenance that will be required.

     2.7.2  Shutdown
            The shutdown of an MWC does not  generally cause significantly increased
     emissions.  Charging is discontinued, but grate movement, airflow, and operation of
     the air pollution control system continue as appropriate until all the refuse in the
     furnace has been burned. Heat retained by the refractory and other interior parts
     keeps temperatures at acceptable levels for the 1 to 2 hours required to completely
     burn out the refuse after charging has been discontinued.

2.8   EQUIPMENT INSPECTION FREQUENCY
             Clean,   efficient  MWC   operation requires  an   aggressive  preventive
     maintenance program.  Such a  program includes getting out of the control room or
     office to periodically inspect, evaluate and repair equipment.
             Table 2-4 presents an  example of a schedule for inspection of incinerator
     equipment by the incinerator staff.  It must be understood that, as an example, the
     inspection frequency  tabulated  here  should  not be interpreted  as  the  only
     acceptable frequency to ensure  reliable  operation of  the system.   For example,
     many of the  hourly and daily inspection checks itemized here may be inverted at a
     specific  facility or  may be  performed on  a "once per shift" basis.  Similarly, in
     some facilities  the  monthly and semi-annual  inspections  may be performed on a
     quarterly basis, especially if the facility routinely shuts down on a quarterly basis
     and the nature of the inspection  requires internal access to the  furnace, ductwork,
     control equipment, etc.
             The  inspection schedule  implies a  certain amount of routine maintenance
     must be performed  as  needed.  For example,  if an hourly check of the conveyors
     indicates a  jam, obviously  it must be  corrected immediately.   The fact that
     different inspection frequencies  are recommended for specific pieces of equipment
     implies that inspections of different levels of detail are required. For example, the
     cooling tower exit temperature  is the only means to measure the performance of
     the cooling water sprays on  an hourly basis. However,  many plants will'check the
     spray  pumps for  excessive  bearing noise on a per shift  basis, and lubricate the
     bearings on  a weekly basis.  Stand-by pumps  are  often run through their transfer
     cycle on a weekly basis. The spray nozzles are usually inspected (and cleaned or
     replaced as needed)  during every scheduled  shutdown.
                                         32

-------
                                  TABLE 2-4

                           EXAMPLE INCINERATOR
                    EQUIPMENT INSPECTION FREQUENCIES
Equipment
Cranes
Feed Hoppers
Grates
Hourly Daily
X
X
X
Weekly
Lube &
Inspect
X
X
X
Semi-
Monthly Annual
X X
X
X
Bi-
Annual
X


Air Box
Conveyors
Refractory
Fans
Rotary Kiln
Diversion Gate
Dump Trucks
Rotary Valve
 Air Locks
Cooling Water
 Sprays
Cooling Chamber
By-Pass Stack
ESP
Stack
Controls &
 Sensors
Quench Pit
 Water Level
x
x
x
x
      See Table 2-3
x          x
X          X
X          X


X          X
                       X
                       X
                       X
                       X
                                             X

                                             X
                                      33

-------
             Differences  in  technology,  design  detail   and  operating   experience
     ultimately dictate required inspection and maintenance  frequencies for specific
     incinerator  components.    More can be  inferred about  system  reliability from
     whether or  not a facility has a written inspection and maintenance schedule that
     they adhere to than can be inferred from the details of such a schedule.

2.9  RECORDKEEPING
             Management  and technical personnel at  MWC facilities require certain
     information about the  plant's  operation so that they  can monitor the  efficiency,
     production  and  reliability  of  the  operation and take  corrective  action when
     necessary.  This information includes data about throughput, operating parameters
     for the furnace, combustion,  and air pollution  control  processes.  Fortunately,
     efficient and reliable MWC operation is totally consistent with good air pollution
     practice.  Therefore, the records and reports maintained by  the MWC  staff are
     normally more than adequate to satisfy recordkeeping and reporting requirements
     that may be desirable as stipulations in air pollution permits. An exception, excess
     emission reporting, is discussed later  in this chapter.

     2.9.1   Furnace and Combustion Data
             Section 3.3 contains an example  furnace and combustion data log.  Such
     logs are maintained  by the furnace  operator.  Typically, these logs are manually
     compiled with readings recorded hourly or bi-hourly.  However, there is a clear
     trend  on newer  installations to compile  this data on computer with current and
     historical data available to  the operator via CRT and/or paper.
             Precisely what data is available and recorded at a  given facility  will .depend
     on the basic  MWC configuration, the sophistication of the automatic  combustion
     control system and the experience/opinions of the MWC management.
              Monitoring the instruments is essential to  maintaining a clean, efficient
     operation and preventing minor process excursions from  becoming substantial .and
     costly  equipment problems and environmental compliance difficulties.  Creating a
     historical record of the process operation provides MWC management with a tool to
     evaluate trends in process operation and malfunctions.

      2.9.2    Stack Gas Monitoring and Excess Emission Reporting
              Figure  2-1 is  an  illustration of  a typical Continuous Monitor  Quarterly
     Report. These reports can  be rather laborious for plant staff to prepare jf. there is
                                         34

-------
                                                CJ
i
o
                                               o

                                          §    I

                                          115
                                          E   1   S
          CO
                                                «=    a
          UJ
       £ Ul
A     1
       ^ O

          CO
                                           «    5    g

                                                     =
                                           c.   i_    ®
                                           a;    o    2
                                           *"    .  "=   "55
                                    =5    a  -s
                                    §.   £    §

                                    ^  ox   §
                                           o.   w   es
                                           o   O    w
                                          v?   CO    O

                                          t   7T  ^
                                          —    c   "5
                                          •a   2    a
                                           0)    tn    O.
                                          ~    to    o
                                          i.  to    o

                                          2-  i    «
                                               es    «
                                          
                                m   »^   >?•   51
                                               —  •  CD
                     •

                                                                 5=  o
                                                                   =  6
                                                                                   35
                                                                                                                                                    C/3

-------
a large quantity of excess opacity excursions during the quarter.  If the facility is
maintaining continuous compliance adequately, the reports can be compiled in a
very few minutes.  It is the  former case that presents a compelling argument  for
permit conditions that require the preparation and submittal of such  reports.
        In  addition  to  documenting  the  time,  duration,  and severity  (percent
opacity) the report requires the  identification of the nature and jcause of the
incident and the corrective action taken. These last two data columns provide the
most  important information in the  report.   Most  often,  those  facilities that
experience difficulty in achieving  continuous compliance do so because they have
one or  two  operating problems that create  the bulk of the violations (Reference
12).  If the  data are summarized  to indicate which problems are causing the most
violations, the plant management and, if necessary, the state or local air pollution
control agency, can focus on  the solution to the problem.
        To  date,   EER's   have   not   been   required  at   MWC's   utilizing
transmissometers.  In view of the wealth of useful information that can be derived
from EER's, it is desirable to re-evaluate this approach.

2.9.3    Miscellaneous Records
        Additional recordkeeping at MWC facilities will vary widely from plant to
plant and,  in general,  will be of marginal use to air pollution authorities.  It is
helpful, however, to be aware of the  broader recordkeeping  activities since some
facilities  will  have useful  environmental data  tucked  away on  a record that
otherwise may seem rather obscure.
        Figures  2-2 and 2-3  illustrate  two forms utilized by the  Montgomery
County (Ohio) Solid Waste Management Division in operating two MWC's at each of
two sites in the  County. Figure 2-2 is the Incinerator Shift Operations Record
filled out at the end of each shift by  the  Shift Foreman.  The record  provides
summary  information  for  the  shift  regarding  operating  time and throughput
(grapples fed and loads of residue  hauled).  Ammonia is used to control pH  of the
process water, so the  amount of  ammonia  used  and the pH at each clarifier is
recorded once per shift at this facility. The ESP is equipped with an oil heater that
is fired during brief shutdowns to prevent acid  condensation.  Thus, the amount of
diesel fuel used per shift is also recorded.
        Figure  2-3  is the Combustion Condition Observation Form.  This form is
used as part of the Division's training program and is used on a random basis to
document operating observations.  The value of the form  is to encourage operators
to make  appropriate  observations of  the  combustion  conditions and, just  as
important, make appropriate  corrections or explain why not.
                                    36

-------
            FIGURE 2-2
INCINERATOR SHIFT OPERATIONS RECORD
                  Plant
Date
Shift No.

Furnace 1
Furnace 2

Shift Foreman
Ooerating Hours
From


To


Truck// Gals
Trucks? Gals

Downtime Hours
From


To

-
Diesel Gals
PH 111 PH #2

Grapples
Fed


Loads of
Residue
Hauled ,

Ammonia Reading
Amount Used


Remarks :



•




              37

-------
                                   FIGURE 2-3
                                                  DATE
                     COMBUSTION CONDITION OBSERVATION FORM

                          Extremely Dry           ,        Very Wet
Condition of Refuse



Condition of Ash #1

Condition of Ash #2



Condition of Plume
    Poor Burnout
       Difty
Drafts:  All Balanced & Approximately Correct:

If not, Explain:
Good Burnout
    Clear
                                  yes
                                               no
Grate & Kiln Speeds:  All Balanced & Appropriate  Conditions: 	 yes        no

If not, Explain:
Combustion Air Fans:  All running, quiet and clean  intake:

If not, Explain: 	
                                           yes
                                                     no
Combustion Air Seals:
filled with water:

If not, Explain: 	
 All  doors shut, rotary locks working, and ash quench
	yes          no.
                                                  Division Chief Engineer
Response of Operator:
Signatures:
Incinerator Operator
    In Charge
        Shift Supervisor
 Chief Operation
    Supervisor
                 Comments                              Plant Manager

(Forward to Superintendent After Reviewed & Signed by All.)
                                      38

-------
2.9.4    Charts and Computers
        Many facilities continuously record furnace and cooling tower temperatures
and opacity on strip or circular charts. Even in such cases the operators are usually
required by management to manually log these parameters.
        The extent to which  computers are  used at MWC's varies greatly from
plant to  plant.   Any  automatic  combustion control system is  a computer,  if
somewhat rudimentary  by today's standards.   Some facilities also  utilize the
computer as a data recorder and  logger.   Operating parameters such as furnace
temperatures  and oxygen concentration can be monitored via a CRT. In addition  to
the instantaneous values, the  computer is usually programmed to provide historical
summaries and hourly averages of  the operating parameters.  Precisely what data
are routinely printed will,  again, vary from plant to plant.  Facilities involved with
steam production usually have the most sophisticated control systems.

2.9.5   Summary
        MWC facilities utilize rather extensive recordkeeping practices to operate
and control  the  combustion  process and  the movement  of municipal waste and
combustion residue  through the plant.  These records are  also useful to the plant
staff in  diagnosing  and resolving  operational problems,  whether  those problems
result in excess air poUution emissions or not. The air poUution control official will
generally  not  need  records  beyond  those  maintained  at  responsible  MWC
operations.   He  should be aware of  the normal (baseline) operating  ranges  of
critical parameters  such  as furnace  temperature,  oxygen concentration, ESP
secondary power levels,  and scrubber pressure  drop that are specific  for  the
faculties in  his jurisdiction.  He should also be aware that operation outside the
baseline  ranges  for these  critical parameters  is  an indication of a potential  air
pollution problem.
        Transmisso meters  have  proven  successful  in  a  number  of  MWC
applications.     Facilities  should  be   encouraged  to   install   and  maintain
 transmissometers that meet the 40 CFR 60 Appendix B specifications in any new or
 existing  facility where  water vapor  in the stack or duct  is not a problem, and, in
 the ease of existing units, where  the remaining useful life of the unit can justify
 the expense.  It  must not be forgotten, however, that the opacity data generated by
 such instruments is totally useless if it is  simply stored in a box at the plant or  the
 air agency's offices.
                                     39

-------
2.10 REFERENCES

     1.      Brunner,  C.R.  (1985).    Hazardous  Air  Emissions  from  Incineration.
             Chapman & Hall, Ltd., New York, New York.

     2.      Air Compliance Inspection Manual.  U.S. EPA Publication No. EPA-340/1-
             85-010, September 1985.

     3.      Richards,  John  R.  and  Robin  Segal.     Wet  Scrubber  Performance
             Evaluation. U.S. EPA Publication No. EPA-340/1-83-022. September 1983.

     4.      Theodore, Louis  and  A.  Buonicore.   (1982).   Air  Pollution  Control
             Equipment: Selection,  Design, Operation and Maintenance.  Prentice-Hall,
             Englewood Cliffs, New Jersey.

     5.      Azabo,  M.F.,  and  Y.M.  Shah.   Inspection Manual for  Evaluation  of
            "Electrostatic  Precrpitator  Performance.  U.S. EPA Publication No.  EPA
             340/1-79-007.  March 1981.

     6.      McDonald,  Jack R.,  and  Alan  H.  Dean.   A Manual for  the  Use  of
             Electrostatic  Precipitators  to  Collect Fly  Ash Particles.   U.S.  EPA
             Publication No. EPA-600/8-80-025. May 1980.

     7.      Control Techniques for Particulate Emissions from Stationary Sources -
             Volumes  1 and 2.  U.S. EPA Publication Nos. EPA-450/3-81-005 a and b.
             September 1982.

     8.      Yung, Shui-Chow, et al. Venturi Scrubber  Performance Model.  U.S.  EPA
             Publication No. EPA-600/2-77/172. August 1977.

     9.      Proceedings: Operation &  Maintenance of Electrostatic Precipitators.  Air
             Pollution Control Association, Pittsburgh, PA. April 1978.

     10.     Proceedings:   Operation  & Maintenance  Procedures  for Gas Cleaning
             Equipment - A Specialty Conference.  Air Pollution Control Association,
             Pittsburgh, PA. April 1980.
                                        40

-------
11.    Kalika,  P.W.   How  Water Recirculation and  Steam  Plumes Influence
       Scrubber Design.  Chemical Engineering. July 28, 1969.  pp. 133-138.

12.    Schmidt, Charles M., and R. D. Allen, "Reducing Opacity Emissions from
       Industrial and Utility Boilers," Paper  No.  84-110.3, presented at the 77th
       Annual Meeting of APCA, San Francisco, CA, June 29, 1984.
                                   41

-------
                                    CHAPTER 3

                  SITE VISITS AND OPERATING VARIABLES STUDY

3.1  INTRODUCTION
             This chapter  discusses  several  operating and  air  pollution compliance
     problems described by MWC operators during the MWC site visits performed under
     this study.   Each "case history"  describes the nature of the problem or concern
     faced by the operator and the action or solution implemented to resolve the issue.
     The kinds of problems addressed in the case histories are fairly typical in the MWC
     industry. Thus, it is hoped that these  presentations may be  useful to other MWC
     operators.
             Also discussed in this chapter are the results of a field study conducted at
     the Stamford, Connecticut MWC. The purpose of this field study was to determine
     the  relationship  of operating  variables  routinely monitored at  the facility  to
     opacity emissions.

3.2  SITE VISITS - CASE HISTORIES
             Case History No. 1 - This facility operates two mass burn rectangular batch
     furnaces rated at 100  tons per day each.  The two cell furnaces are equipped with
     rocking  grates.   The  furnace  exhaust  gases are  treated by a single  two  field
     electrostatic precipitator.  The facility operates on a five day per week schedule.
             The  ESP  was  installed in 1980  as part  of a  refurbishing and  upgrading
     program for the  overall system.  The  operating  staff and engineering contractor
     were aware of the corrosion difficulties ESP's suffer in MWC  service. Further, the
     facility's five day per  week operating schedule  was  expected to exacerbate the
     corrosion potential since each shutdown  would cool gases in  the ESP to below the
     acid dewpoint.
             To eliminate this problem, an oil heater was installed in the inlet duct  of
     the  ESP.    The  heater  is  fired during  weekend shutdowns to  prevent  acid
     condensation.  The oil  burner  consumes  200  to 300 gallons of fuel  oil during the
     weekend to  maintain  ESP internal temperatures around  275°F.  Often it  is not
     necessary to fire  the burner in the summer.
             The ESP's are inspected internally once  per  month.  Corrosion has been
     negligible over the seven years of operation.  The facility superintendent  feels that
     the cost of  the fuel oil is insignificant compared to the benefits derived from the
                                         42

-------
burner. However, in view of current fuel oil costs, he would recommend a burner
that utilizes natural gas.
        Case History No. 2 - This facility was experiencing fugitive dust problems
resulting from tipping floors that were not fully enclosed and unpaved roadways on
the plant premises.  In 1982 the tipping areas were completely enclosed with sheet
metal siding and large overhead garage doors. Further, a program to install asphalt
paving on unpaved roads was initiated  in 1981 and completed within three years.
These newly  paved  roads are  swept  regularly with  automatic road sweeping
equipment.   This program has brought the facility into compliance  with  local
fugitive dust control'requirements  and  has resulted in a substantial reduction in
citizen complaints.
        Case History  No. 3 - Exhaust gases from four rotary kilns are treated by
each of four venturi (cone-type) scrubbers.  The scrubber liquor is a 100 percent
recycle: make-up is for evaporative losses only.  Liquor treatment is provided by a
clarifier/still well system. No provision  is made for pH control.
        The system was  suffering from  persistant  and advanced corrosion in the
scrubbers, mist eliminators, ducts, and ID fan housing.  Observations at the site of
stack gas emissions confirms opacity measurements varying from 20 percent  to 80
percent during  normal operations.   Furthermore,  little  difference in the plume
appearance could be noted with and without the venturi scrubbers  in service.  This
suggests a plume  comprised predominately  of  submicron particles that are not
removed  by the scrubber.   Such  particles are typical of  the  soot and cracked
hydrocarbon tars which characterize opacity due to incomplete combustion.
        At  the time  of the site  visit, the MWC management had  installed an
ammonia system to control the pH of the scrubbing liquor.  In  addition, they had
begun a program to refurbish each of the four furnaces and  control systems one at
a time. The refurbishment of the first  system was near completion at the time of
the  visit.  The project  had included  rebuilding  the ducts,  scrubber and mist
 eliminator to  replace corroded parts  and repair holes and cracks as necessary.
 Additional water  capacity  was added to the  presaturator. To improve combustion
 conditions, overfire air was added to the furnace.
        Case  History No. 4. - The  ESP  ash  hoppers  at this MWC were suffering
 frequent bridging problems.  The original installation of the system did not include
 vibrators on the hoppers to  aid ash removal.  Installation  of vibrators solved  the
 problem.
                                    43

-------
        Case History No. 5 - This facility consists of two rotary kiln MWG's. As is
typical  with such systems, the kiln is preceded by a furnace equipped with drying
and ignition grates and combustion air supplied by over fire and under fire nozzles.
Air pollution control is achieved  by two ESP's with three fields each that vent into
a common stack.   The  ESP's were  installed  in 1982, replacing a wet scrubbing
system.
        During the first year of operation the ESP's experienced  re-entrainment
problems when the third fields  were rapped, causing opacity violations. The re-
entrainment  problem  did not occur  everytime the third fields were rapped,  but
appeared to occur only when poor combustion occurred in the furnace, specifically
when the furnace was burning highly combustible waste.  The resulting incomplete
combustion produces fine, light particles of soot that the ESP's can collect, but due
to the low  resistivity of the particles, they become re-entrained during rapping.
        To resolve the problem, the frequency of third field rapping was reduced to
once every 24 to 48 hours and interconnect ducts were installed between each MWC
line and ESP. This allows the exhaust gases of both MWC's to be vented to one ESP
while the third field on the idle ESP is rapped.  With no airflow in the idle ESP, no
opacity violations can occur during rapping.
        After installation of the interconnect system,  opacity violations were
reduced to 1.67  percent  of the total operating time. Almost half of the remaining
violations  were  related  to  startup  of the  facility.   Yet, the larger half still
appeared  to be  related  to  some kind of uncontrolled combustion caused by  a
specific waste.   By  coordination of various  members of  the  operating staff the
problem was isolated and identified as a highly combustible waste from a styrofoam
processing plant.   Apparently,  the resistivity/re-entrainment problem  related to
this waste was so  severe, the  ESP's  simply  could not handle it, even with the
interconnect  procedure.   The firm  was contacted and informed  that the MWC
facility could no longer accept the firm's waste.
        Case History No. 6 - In 1983 this facility shutdown  two furnaces to perform
an  extensive rebuilding of  the  ESP's  serving the furnaces.   The rebuilds were
necessary  because of extensive corrosion in the ESP's, particularly in the outboard
plates where holes  of several  square feet  in  area had developed.  Prior to the
rebuilding  effort the ESP's had operated reliably for almost nine years. The facility
superintendent  felt  that  this is good performance for  ESP's  in  MWC  service.
Nevertheless, the superintendent is taking precautions to ensure that the MWC
system is  operated  in a  manner  consistent with long term  reliability of the rebuilt
                                    44

-------
unit.    This  includes  maintaining furnace  exit  temperatures  and ESP  inlet
temperatures  as  high  as possible  to   minimize  the chance of condensation and
carryover of cooling tower spray in the ESP.  For  these units, maximum furnace
exit temperatures are  1600°-1650°F.  Temperatures over 1700°F tend  to cause
slagging.  The ESP maximum inlet temperature is set at 650°F. The units  have also
had some difficulty with welds breaking in the hopper assembly, allowing air to leak
in.  These breaks are repaired during scheduled downtime.
        Case History No. 7 - This facility consists of two RDF fired traveling grate
stokers with tail end boilers. Emissions are controlled by ESP's.  Local air pollution
officials had documented several opacity  violations and had received complaints
about ash fallout in the neighborhood.
        Although  the combustion air for this system is controlled automatically,
the feed rate, as dictated by the speed  of the screw conveyor, was controlled
manually.  The MWC staff developed software  to provide automatic control of the
feed  rate  based  on  the  steam  demand.   Automatic  feed control. eliminated
quenching the bed by feeding too much fuel.  To  gain further control of the excess
air  rate,  the  staff  repaired cracks in the furnace system to minimize  air in-
leakage. The  combination of these measures eliminated the opacity and ash falldut
problems.
        Additional system refinements are  planned for the future.  First, the
operators are planning  to  install, oxygen and  carbon monoxide monitors.  This
installation is  part of a long  term plan to provide combustion control through these
parameters rather than  furnace  temperature, which is the current practice. The
operators also believe they can provide better combustion control by eliminating
the fly ash reinjection system.   Finally,  the ESP's pneumatic rappers will  be
replaced by hammer  rappers.  It is believed that  the pneumatic rappers  do not
provide sufficient energy to properly clean  the plates.
        Case History No. 8 - Two ESP's control particulate matter emissions from
two travelling grate RDF fired furnaces. Auxiliary fuel is used during startup and
when flame out occurs on the fuel .bed.  Originally No. 6 oil  was the auxiliary fuel.
The oil created  opacity  violations (the  standard is 20 percent)  and  citizen's
complaints  were  registered.   These problems were  eliminated by switching to
natural gas.
        Another problem at  this site  has  been clinker  formation in the ESP ash
hoppers. The  clinker causes  the ash to bridge, thus backing the ash up into the ESP
itself.  When the ash approaches the plates the electric field is disrupted, causing
                                   45

-------
     the ESP to short  out.  The MWC staff has investigated the problem and they are
     convinced  that the bridging is not caused by air in-leakage/condensation.  At the
     time  of our  visit  they  were considering rebuilding the hoppers so that the hopper
     walls  are  almost  vertical  and the  openings to the  ash conveyors below  are
     substantially larger than the 8-inch diameter valves currently in service.
             The fuel delivery system (air blast distributors) has suffered some abrasion
     problems in  the  ducts where the RDF is blown onto the traveling grate.  The
     abrasion apparently is caused by glass contained in the RDF.  (Refuse processing at
     this facility does not include an air classification system that would eliminate glass
     from  the fuel. Basically, the fuel processing at this facility  entails hand sorting,
     magnetic separation, and  shredding by a hammer mill.)  Although the problem has
     not been completely  eliminated, some improvement  was noted when a local bottle
     bill went into effect.  Returnable glass bottles resulted in less glass in the RDF and
     also resulted in a small but noticeable increase in the heat content per unit mass of
     the RDF.

3.3  OPERATING VARIABLES VS. OPACITY STUDY
             In  an effort to determine how  MWC operating practices affect visible
     emissions,  a  field study was initiated in the City of Stamford, Connecticut at their
     co-disposal (municipal  solid waste and sewage sludge) incineration facility.  The
     purpose of the study  was to attempt  to correlate "real world" operating  data
     routinely compiled by the MWC  staff to the  visible emissions emitted  by  the
     facility.  The operating parameters  such  as furnace temperature,  under fire air
     pressure, furnace pressure, etc. recorded hourly over ninety days were statistically
     correlated to the hourly opacity readings from the opacity monitor.
             The 360 ton per day (tpd) Stamford MWC was chosen because the facility
     met several selection criteria established by the EPA. Specifically,

     1.      The facility has a good overall visible emission compliance record and is in
             generally good  mechanical condition.  Therefore, it  was known that good
             operating practices can produce good results  at this facility.
     2.      The Stamford staff has conducted several studies on combustion efficiency
             and throughput and thus has experience in data collection and analysis.
     3.      Except for the  co-firing of sewage sludge, the equipment configuration at
             Stamford  is  reasonably  typical  of  intermediate to large  size  MWC's
             currently in operation in the United States.
     4.      The Stamford facility has an opacity monitor.
                                         46

-------
5.      An east coast location for the test  facility was necessary for economical
        travel for EPA Headquarters and contractor staff.

        This section provides a process description of the Stamford 360 tpd MWC.
This is followed by a  description of the field study protocol.   Finally, a brief
discussion of the results and conclusions are presented.
3.3.1    Process Description
        The 360 tpd MWC in all ways, except one, is a "typical" mass-burn unit with
continuous feed of unprocessed municipal waste, rocking grates,  and an ESP  to
treat exhaust gases.   The  one exception to this  unit being typical is the use  of
energy to process the solids removed from the sewage  treatment plant. The MWC
is designed to process  30 tons per day  of dried sewage sludge with  a maximum
moisture content of 20 percent from belt filter presses.  The sludge comes  from the
city's 20 million gallon per day  (mgd)  wastewater treatment plant located on the
same site.  Sludge processing is accomplished by two  belt filter presses, a rotary
kiln that uses  a slip  stream from the  furnace  exhaust to dry the  sludge, and the
MWC.  The  dried  sludge is fed to the MWC continuously from the  kiln through
several openings in the furnace ceiling. Figure 3-1 is a schematic of the Stamford
waste processing facility.  Flyash laden water  from the process water system is
pumped to a separator/clarifier that is part of the MWC. The city also operates a
150 ton per day batch MWC at this site that is not a part of this study.
        Waste  Handling and Storage - Trucks dump refuse into a receiving bin that
has a capacity of 425 tons level fuU and 725 tons heaped.  This design capacity is
based on the assumption  of a refuse density of  12.4 pounds per cubic  foot.  (More
typically, refuse density is 25 lbs/ft^ which means that the receiving  bin capacity
may  hold  twice  the  design  capacity).    The  refuse consists  of  residential,
commercial, and industrial waste.  Other  types of waste such as  pathological or
hazardous  wastes are not processed at  this faciliity. The proximate analysis of the
municipal  refuse burned is typically 55 percent combustibles, 30 percent moisture,
15 percent ash and has a heat content of approximately 5400 Btu/lb.  The furnace
feed  hopper is served by two air conditioned cab  cranes each with a  2-cubic yard
grapple.   The hopper delivers refuse to a refractory  lined continuous feed chute
connected to the furnace.  Refuse in  the  chute provides an air-tight seal for the
furnace when  in operation.  A  hydraulically powered cover seals off the hopper
during shutdown periods.
                                    47

-------
                                          Figure 3-1
                             Schematic of Stamford Incinerator
                                          PRECIPITATOR
                                         225,000 CFM at 600°
    PRECIPITATOR
  750,000 CFM at 600'
                       	           *rr-	—   »  f
                      HOPPER^ /SLUDGE INJECTION NOZZLES I
                                                                             CENTRIFUGALS
                                                                          EXHAUST SILENCER
                             II  -   -
                            UNDERF1REOUCT
           UNOERFIREFAN
                                                                                       "ARY SLUDGE DRYER
OVERFIRE DUCT j  QVERFIRE   ,	,
                "/FAN   CHEMICAL MIXING TANKS
                                                                                         lENCY GENERATOR
                                SLUDGE RECEIVING
                                      WELLS
Source: Reference 1
                                              48

-------
       The receiving bin is equipped with water sprays that are designed primarily
for fire  suppression.    When  sensors  in the  bin  roof detect  abnormally high
temperatures, alarm annunciators in the  control room signal  for manual operation
of the spray nozzles.
       Three sludge wells are located  in the building in addition to sludge pumps,
belt filter presses, drying equipment,  and chemical conditioning storage/mixing
facilities.  With a maximum  moisture content of 20 percent,  the sludge burns at a
maximum hourly rate of 1.25 tons.
       Hot gases from the  furnace are used in a sludge drying heat exchanger
(rotary kiln) with heat exchange exhaust ducted back to the combustion chamber.
       Dried sludge is  introduced into  the furnace through openings in the furnace
roof,  with burning taking place in  suspension prior  to  reaching  the  furnace bed.
Typically,  the dried sludge fed to the  furnace will have a heating value of 3000
Btu/lb and a moisture content less than  10 percent.
       Furnace and Grates  - The  combustion chamber is of wet-bottom design;
i.e., a poor of water is maintained in the bottom into which bottom ash and larger
particles in the exhaust stream settle  out and are removed  in a slurry.  A water
make-up system  automatically maintains the desired water level with manually
operated dump valves provided for the draining of the flyash laden slurry.
       The lower furnace walls are subjected to abrasion  from moving refuse.
Thus, this  portion of  the furnace  is  outfitted with air cooled silicon  carbide
refractory materials.
        Waste is  agitated and transported through  the  furnace by  a series of
rocking grates on which the refuse burns.  Rocking grates  are  arranged  in rows
across the width  of the furnace, at right  angles to the solid waste flow. Alternate
rows  are mechanically pivoted or rocked  to produce an upward and forward motion,
thus  advancing  and  agitating the  solid waste.   The rocking  grate stoker is
hydraulically  driven.  The same hydraulic system  also operates the  refuse hopper
cover and a chute at the high end of the  ash drag-out conveyor.  The furnace grate
area  is 355 ft2 with a grate loading of  approximately  80  lb/hr/ft2.  The  furnace
dimensions are 12 ft. in width, 39 ft. in  length, and 29 ft. in height.
        Combustion Air - Combustion air is  provided  by underfire air, sidewall
cooling, and overfire air forced draft systems.  Underfire air is that  portion of  the
combustion air which is introduced to  the chamber from beneath the grates. The
underfire air  is manifolded to several areas under the grates, where  it is directed
upward so that the air passes through the  grates and refuse.
                                    49

-------
        The underfire air is furnished by a 70,000 cfm fan that is modulated by an
inlet  damper.  The  underfire air fans are designed to provide sufficient velocity
(energy) to the underfire air jets so that the combustion air penetrates the fuel bed,
thus providing maximum contact  of  combustion air  with the solid, combustible
material in the refuse.
        In   the  automatic   mode,   the  volume  of  air  delivered  by  the
underfire/overfire system is  regulated  through a control" system  driven  by a
temperature sensor located near  the furnace ceiling. The control system increases
the volume of underfire air  if the furnace temperature  is too low simultaneously
reducing the over fire air.  The volume of underfire air is decreased if the furnace
temperature is too high, simultaneously increasing the overfire air.  The movement
of the  underfire/overfire air dampers is always proportional, i.e., they move in
opposite and approximately equal amounts. By viewing Figure 3-2 one can see the
position that the underfire and overfire are at in respect to each other both inside
and outside the selected burning temperature. The amount of air being delivered to
the furnace by the overfire air system  is monitored by a draft gauge on the overfire
air duct just after the overfire air fans, the amount delivered by  the  underfire air
system  is  monitored by a draft gauge  on the  underfire air duct just  after the
underfire air fans, and the total air movement through the incinerator is monitored
by a draft gauge on the outlet side of the forced draft fan.
        Two 12,500  cfm fans with inlet damper  controls supply overfire air. The
reasons for  introducing overfire  air are to ensure,  through added turbulence, the
complete oxidation of combustible gases that have evolved from the solid refuse
fuel bed, and to control high  temperatures in the furnace. (As described above, an
increase in  the overfire airflow quenches the  furnace gases.)   Overfire  air is
introduced above  the  fuel bed through the furnace sidewaUs.  The overfire air
system was designed for complete penetration of the exhaust gases.
        SidewaU cooling is supplied by two 15,000 cfm fans.  Although the primary
purpose of  sidewall  cooling  is  to protect  the refractory,  sidewall cooling air
contributes to the overfire combustion air. The sidewall cooling air fan damper is
manually operated and normally set at  40 percent open.
                                   50

-------
   CD  £
   CD
  CD

  .9-E
  JZ  0
   O
CD .—
13 ^

D) CD
  ±o
   CC

   D
CD
   o-t:
   CD  0

   SI
< \
I <
1
1 
-------
        Air  Pollution Control - The  emission  control system  consists of a spray
chamber followed by an ESP. The water spray chamber reduces the combustion gas
temperature to between 550°  and 650°F   prior to entering the ESP.  The spray
chamber consists of a  multiple  bank spray  system,  with cluster  type nozzles
providing 100 micron diameter droplets at 125 psi.  Spray volume is automatically
controlled in 45  to 50 gpm increments.  The spray chamber has high-heat  heavy
duty  fireclay  brick  wall§  with  a  castable  refractory  similar  to  that of  the
combustion  chamber.
        The ESP is located  on the roof because  of  site restrictions.   The ESP
consists  of  two electrical fields and  is designed to handle 225,000 acfm at  600°F
and a maximum pressure drop of 0.5  inches water  column.  Motorized dampers on
the inlet duct  control flow under abnormal operating  conditions.  The  cross-
sectional area  of the ESP was  designed to provide a velocity of 3 to 4 feet per
second through the unit in order to allow adequate detention time and limit dust re-
entrainment.  ESP  design parameters are summarized in Table 3-1.  An induced
draft  (ID) fan  is downstream  from the ESP.   The fan has a radial'tip blade, a
variable speed drive unit, and is sized  for 202 bhp.
        Inlet dust loading conditions are 3.5 pounds of particulate matter per 1000
pounds of dry  gas, adjusted to 50 percent excess  air for product of combustion.
Maximum particulate matter concentration in the ESP discharge is 0.175 pound per
1000  pounds of dry gas  was  specified;  equivalent to 95  percent  efficiency.
Sampling ports  are  installed  on  the inlet  and  outlet ducts  of  the ESP  for
performance testing.  Sampling ports are also available on  the exhaust stack.
        Multiple  hoppers  at the bottom of the ESP collect  flyash, with rotary
valves for removal.  Slide gates have been installed upstream of these valves to
facilitate maintenance without emptying the hoppers.  From the rotary valves, the
collected flyash drops into a sluicing system and is conveyed by suspension in  water
to the ash drag-out conveyor trough.
        The  ash  handling  system  consists of  stationary packers  receiving  the
residue  from the ash  drag-out conveyor.  The residue is then transferred into  ash
receiving containers and hauled to  disposal sites.
                                   52

-------
                                    TABLE 3-1
                                ESP PARAMETERS
Equipment Parameter
Control Efficiency (Design)
Plates
No. of Fields
Wires
Plate Area
Distance from wire to plate
Rappers

Hoppers

Instrumentation
Value
95% minimum
18 gauge cold rolled mild steel
2
Weighted Wire Design
39,000 sq ft
4.5 in
Solenoid (Replaced  later  with pneumatic
rappers)
60°  hopper  slope,  electric  heaters  and
vibrators. Discharge valves: rotary feeders
Analog meters for primary  and  secondary
current and voltage for each field.  Spark
meter on inlet field only.
Flue Gas Parameters

Air Volume (design)
Gas Velocity

Inlet Temperature
Maximum Temperature
Exhaust Gas Moisture
Inlet Particulate Loading
Outlet Particulate Loading
Value

225,000 acfm at 600°F
3.0 ft/sec minimum
4.0 ft/sec maximum
550°F - 650°F
700°F
1500 Ib/min
3.5 lb/1000 Ib1
0.175 lb/1000 Ib1
 '•Pounds particulate per 1,000 pounds dry gas @ 50% excess air.
                                       53

-------
        Instrumentation Controls - An air-conditioned control room is located on
 the  ground floor  near the discharge end  of  the furnace.   The control room  is
 furnished  with basic  instrumentation  and controls for the furnace  and ESP.
 Automatic  controls  and  panel readouts are  provided  for  air  and water  supply
 pressures and  flow  rates.  Fans, pumps, and  control dampers are monitored and
 controlled manually when not  in  the  automatic mode.  The temperature probe
 measuring furnace  exit temperature is located  near  the center of the primary
 chamber  ceiling.  AU  the temperature probes are shielded to prevent false high
 readings attributable to radiant energy within the  combustion chamber. Continuous
 gas sampling and analysis for smoke density is conducted by a self contained system
 located on the  third floor.  Operating procedures require daily calibration.
        An opacity monitor is installed on the  exhaust stack. A chart recorder for
 data storage is included with the opacity monitoring system and is located in the
 control room.   Records were not available to indicate if  the  unit has passed a
 performance specifications  test or when the  last relative accuracy  test  was
 conducted.  The instrument zero and span drift is checked and adjusted daily by a
 certified instrumentation technician.

 3.3.2   Field Study Protocol
        The field  study was  a cooperative  effort between the City of Stamford's
 Bureau of Sanitation and the U.S. Environmental Protection Agency's Office of Air
 Quality Planning and Standards.  After review and acceptance  of  the field study
 protocol by each of the  parties, the Bureau of Sanitation  was responsible for
 coUecting approximately 60 days of data on process and operating conditions, and
 visible emissions.
        Figure  3-3 illustrates the  operating parameters provided by the City of
Stamford.  All  the data except  oxygen, carbon monoxide and carbon dioxide were
part of the original study concept.  The gas concentration data were coincidentaUy
being coUected by the MWC staff as  part of a  preliminary  program to look at
certain conditions to assess the feasibility of: 1) if and what should be replaced of
the  antiquated  nitrogen  dioxide,  oxygen,  carbon  monoxide  and  carbon dioxide
monitoring equipment supplied with the facility in  1973; and 2) the possible addition
into  the over fire/under fire/temperature control  loop of feed with trim control
within  span provided by the carbon monoxide concentration.  To minimize expenses
for the preliminary program, the Sanitation Department borrowed a CO/CO2/O2
                                   54

-------
Q*
             r
             g
             H
             G.
             -H



             i
             0
             «J
u
tl
CA

V
X.
0
hi

to

•a
o
o
CM
                     Q 4)
                     E 41
                     4) U
                     •U O*
                       U
        a
        H

        &
        U
        4>
     «*>
 10
 11
 12
 13
 IS
«r¥j
      Vf

      u->
  17
  ia
  19
  20
  21
  23


            / •» f;'
?*<>'





                     «*
             A3

                      *?,.;
                          ts
                          tffi:

condition

-dry/vet
Hall
Furnace pre»«ure
                                       ;2
                                      - II
                                \~\f-.
              '/

             **,
                              Vf
                          /O
Smoke »eter rending
Through put/hr
                                          Iv'W&o
                                      'it
                                       •tl
                          • n
                           h
                           in
                                       ta
                                   n
                                                 n
                                                  -X
                                           7.3
                                           7s
Dioxide
Carbo
S
E
S
S
                                               £35_J_W?/
                                                              •/c
                                                                            M
                                                              J*<
                                                                  Ml
                                                           z^a.i
                              it.?
                                         rf",
                       /o  /y£
                           /t)
                                        li
                           ia
                              'f*.
                                    /?
                                                         ?>
                                           17*
                                                  t.o

                                                               •r /
                                                                        i^v/
                                                                            '  -/
                                                           ,i/  A » | ,?/
                                                            IM
                                        •:/
                                     11..
                                       .w
                                                         ?'H
                                               ,'4^
                                                               1.-1
                                               1,3.
                                                                    *
                                                      ?fl
                                                    .0 V//
                                                          ;r I «s
                                                             f>
                                                                         l,7«
                                                                              '-
                                                                511,24
                                                                     3Ll,n
                                                                IW
                                                            Kl
                                                               V t
                 1)   Peed stroke • add  strokes/sect ion  x 12 •«•  s«c delay ti:
        [
        \
                      7 feed  section  setting
                      30.67
                                               27  x 250  •  lb« per hour
ncinerator
                                                                                      T3
                                                                                      U
                                                                                      O
                                                                                      UJ
                                                                                      S
                                                                                      
-------
 instrument that was designed to compute combustion efficiency for oil, gas or coal
 combustion.

        During the course of the study, it was desirable to  manipulate some of the
 operating variables over  specific ranges to broaden the data base.  Specifically, it
 was agreed to attempt the following:

                  Furnace Temperature:   During the course of the study, it was
        recommended that the furnace exit  temperature be run from  1600°F to
        1900°F  and back to 1600°F in 50 degree increments. To ensure  that the
        furnace stabilized at each temperature level, each increment  was to  be
        held for  at least  four  hours.   After  considerable discussion  with the
        Superintendent of  Sanitation, Mr.  Marvin  Serra,  and  the  Solid  Waste
        Division Staff, it was decided that determination of upper and lower bounds
        through opacity violation would not be  an acceptable means.  Further  it
        was determined that 1) variation of the temperature, because it affected
        throughput, could only  be accomplished  when time permitted,  and  2)
        operating above 1900 degrees F could not be permitted because of possible
        high heat problems with the ESP.

               Corona  Power:  The corona  power  would be  adjusted  through
        controlling the voltage.  The run  would be  similar to that done for the
        furnace temperature.   However, some  experimentation  was required  to
        determine  how  low the power  could  be  set without  causing  opacity
        violations.  At the upper end, an excessive spark rate could be noted. Once
        the upper and lower  limits were defined, the test would begin  at the lower
        power setting and increased  incrementally every four hours to  the upper
        power limit and back to the lower limit.

        Manipulation of other  operating variables was  considered but rejected  as
not being practical.  For example,  fly ash resistivity is  dependent on flue gas

temperature (among  other  things).   The  range  of responsible  ESP  operating
temperatures is not very large.  The  upper limit is defined by what the unit can
take without suffering heat damage.  The  lower limit is defined by how  much

water can be injected  at the  cooling tower without carryover of liquid droplets into
the ESP.  Since this temperature range is limited, the fact that the temperature
would not be controlled to better than plus or minus 10  to 20°F became significant
to the overall temperature range available.


3.3.3    Discussion, Results, and Conclusions

        The operating  parameters were run  through a multiple regression analysis
to determine which parameters had  a  good  correlation  with opacity.   It  was
anticipated  that several parameters could provide a good  correlation, indicating

that monitoring and controlling certain process operating variables would produce
acceptable emission levels.
                                   56

-------
        Operating personnel were  only  partially  successful in regulating furnace
temperature and ESP corona power and thus determining their impact on  stack
opacity. The reason that the furnace temperature could not be easily regulated can
be attributed to several factors.   The two most  significant factors are the  broad
range  of refuse burned and the  inability of  the  system  to  control  the  refuse
through-put (the majority of the time it was required to be at maximum).
        One of the methods most commonly used  to analyze data is a simple  linear
regression and correlation, from  which a correlation coefficient can be derived.
The  correlation coefficient is an index measure of the degree of linear  association
between dependent and independent variables.  The linear regression equation is the
straight line that best fits the data. The closer the  correlation coefficient is to
unity (1.00), the stronger  the linear relationship  is likely to be. Correspondingly,
for inverse relationships, the strongest correlation coefficient would be -1.00.
        Conclusions - The following conclusions were drawn from the analyses.
1.      The  statistical analyses  that  were  performed emphasized opacity  and
        carbon monoxide concentration in the exhaust gases as dependent variables,
        while other  process  parameters  such as  carbon dioxide  concentration,
        oxygen  concentration,  furnace throughput,  furnace  temperature, ESP
        temperature,  etc., were  considered  independent variables.  The  analysis
        involved determination of the arithmetic mean and standard deviation of
        the  variables, the  correlation coefficient to  measure the strength of
        relationship between two variables, and a simple regression equation (which
        shows the best aggregate correlation between variables).
2.      Statistical analysis of the  available  data was unable to  establish any
        reliable  correlation between  stack  opacity  and  any  other independent
        variable.
3.      Analysis  of  the  data  also showed  no  strong correlation between CO
        concentration  and any  variables  such  as furnace  temperature (-0.12),
        carbon  dioxide (-0.05), oxygen (0.14), refuse burn rate (0.07), ESP inlet
        temperature  (0.05),  or   stack opacity  (0.03).   The  low  correlation
        coefficients  among the gas concentrations may be the result of utilizing a
        gas monitoring instrument that was not designed for the use employed by
        this study.
 4.       All of  the outputs (meters and strip charts) indicated that the variables
         were constantly varying at various rates.  This is typical and expected for
        the operating  variables at an  MWC.  Recording instantaneous readings at
                                     57

-------
 the  end of  each  hourly period  may raise  the  question as  to  how
 representative the  data may be of the actual operation for that hour.
 Ideally,  the pluses and minuses associated with each reading  would, over a
 90 day period, cancel each other out.  The correlation factors resulting
 from this analysis indicate strongly that 1) the sample population may not
 have been large enough, or 2) the high variability in refuse composition and
 operating conditions had a strong impact on the analytical results.  If the
 latter is true, an automated data logger that can output hourly averages of
 the  dependent   and  independent variables,  rather than   instantaneous
 readings at the  end of each  hour, may have  produced  higher correlation
 coefficients.  However, installation of such a device would  have violated
 the "real world"  operating and recordkeeping  criteria defined in the  first
 paragraph of Section 3.3.
 Another difficulty  in  analyzing MWC operating data is  the  lag time
 between cause and effect when adjusting one operating variable to improve
 the value  of another.  For  example, if  the  combustion air is adjusted
 because  the  furnace temperature moves out of  the acceptable range,  the
 furnace  temperature will not  respond instantaneously, but only after some
 period of time -  perhaps anywhere from several minutes to one-half hour,
 or so.
 The opacity  monitor was  audited on July 24,  1985.  The monitor is of a
 design common in the 1960's and early 1970's where an automobile spotlight
 is used as the light  source, and an unfiltered selenium photocell  is used as
 the detector.  The monitor meets the requirements of the operating permit
 issued by the Connecticut Department of Environmental Protection.  The
 U.S.  EPA  does  not require opacity  monitors at  municipal  incinerators.
 However, monitors of  this  design are not capable of meeting the  U.S. EPA
 performance specifications contained in 40 CFR 60 Appendix  B.  However,
 the audit report  (Appendix A) concluded that  the  unit's performance  was
 adequate for  the purpose  of  the Stamford pilot  study.  In addition to
 summarizing the  audit results, the  report  provided  recommendation  for
 calibration and upgrading the system.  The audit was originally scheduled to
 coincide  with the start of the pilot  study.   However,  due to process
 equipment  replacement  requirements for the  study  (ESP control system
replacement  which  was completed before  the data collection), and  the
replacement of the  opacity monitor (which was not completed before the
                            58

-------
3.4
1.
data collection) the pilot study was delayed.  Startup of the pilot program
actually occurred on March 31, 1986.

REFERENCES
Gates, D.W.,  Incinerator is  Part of Integrated Waste  Disposal System,
reprint from Public Works Magazine, May 1974.
                                     59

-------
           APPENDIX A








              AUDIT




         OPACITY METER




STAMFORD MUNICIPAL INCINERATOR

-------

-------
             AUDIT

         OPACITY METER

STAMFORD MUNICIPAL INCINERATOR
              BY

       THOMAS H. ROSE


        JULY 26, 1985
 EASTERN TECHNICAL ASSOCIATES
          BOX 58495
     RALEIGH, NC  27658

-------

-------
                            TABLE OF CONTENTS
I.    INTRODUCTION




II.   TRANSMISSOMETER SYSTEM DESCRIPTION




III.  AUDIT PROCEDURES




IV.  ' AUDIT RESULTS




V.    RECOMMENDATIONS FOR CALIBRATION/OPERATION OF PRESENT SYSTEM




VI.   LIGHT SOURCE AND CONTROL SYSTEM




VII.  RECOMMENDATION FOR REPLACEMENT OF PRESENT SYSTEM

-------

-------
I. INTRODUCTION

   Eastern Technical Associates was contracted to perform a survey audit of the
   transmissometer located on the municipal incinerator in Stamford, Ct. The
   purpose of this survey was to determine the suitability of the transmissometer
   in determining the effects on in-stack opacity by varying the operational
   parameters of this incinerator.  The incinerator is a well maintained and
   operated unit equipped with an electrostatic precipitator. The unit has a
   history of operating within the emission limitations of the State and EPA.
   The transmissometer is on the incinerator stack for operational controls only.
   This source is not required to have an opacity meter. Thus the transmissometer
   is not required to meet the EPA specifications for a continuous emission
   monitor. The study of the effects of changes in incinerator and collector
   operational parameters on in-stack opacity does not require .cull compliance
   with SPA GEM specifications.

   The management and staff of the municipal incinerator were extremely helpful
   in assisting in the conduct of this audit to the point of working overtime to
   assure our success.  Their assistance and cooperation is sincerely
   appreciated.

-------
II. TRANSKISSOMETER SYSTEM DESCRIPTION
A. GENERAL SYSTEM DESCRIPTION

   Tlie transmissometer system  is  very  similar  in  design  and  operation to those
   used in smoke generators  used  for opacity training  and  certification in the
   late 60's and early 70' s. It is  a single pass  unit  consisting of a automobile
   Searchlight source, a  selenium photocell and a stripchart recorder.

B. INSTALLATION

   The transmissonieter is installed on the stack  just  above  the roof over the 4th
   floor of the incinerator  building.   The stack  diameter  at that  point is
   approximately 10  feet. The  stripchart  recorder and  control console is located
   remotely in the main incinerator control room.

C. SOURCE

   The source  is a 6.3 volt  automobile spotlight that is  controlled via a variac
   and stepdown transformer.   This  allows the  adjustment of  the light intensity
   to match the range of  the photocell.  As the  light  intensity goes down the
   beam becomes rich in infrared  radiation and the system  becomes  insensitive to
   the smaller particles  responsible  for the  scattering of light most visible by
   human observers.  Thus  this  type  system is  almost never  photonic and  generally
   underestimates opacity.   The  type  of light  used has two distinct beam widths.
   There is a  stronger inner beam and  a wider  weaker beam.  This is not really
   suitable for a transmissoineter .

D. DETECTOR

   The detector  is a selenium  photocell probably manufactured by International
   Rectifier and  repackaged. It  is  unfilterecl  and therefore not  fully  photopic  in
   response. This  type of cell can  be linear  if operated with an output of over
   15 railli volts. At the  time  of the survey  the sensitivity or output  of  the cell
   was attenuated  to 20 millivolts  with a 1  turn 20 1C Ohms pot. The wiper ^ of the
   pot was worn  out  and  it was difficult to  adjust and was not stable.  With tne
   assistance  of  the staff electrician we replaced the pot with  a  Burns 10  turn
   precision  pot  of  equivalent resistance.   This corrected the instability
   problem and made  the adjustment  of the output simpler.

E. RECORDER

   The recorder  appeared  to be in good condition and responded to  values  across
    the scale.  Due to the  nature of  the source and the detector the chart  was
   difficult  to  read.  The quick response time of the system can  be reduced  to
    alleviate  this problem.  The recorder is set  up to report in Ringleman
    •lumbers,  from 0 to 5.

-------
III.  AUDIT PROCEDURES
A. STANDARDIZATION

   The normal standardization procedure  (calibration) could not be performed.
   This was due to the fact that  it  is a single pass instrument and the
   incinerator could not be turned off.  Instead we performed a relative
   calibration on the system.  Using' the average opacity  from visible emissions
   observations conducted on the  stack the baseline was established.  Normal
   procedure would be to establish this  as zero on a clear stack.  The span value
   or slope was established by blocking  the light source  off .This is in
   accordance with established practice.

B. LINEARITY

   Normal linearity tests are performed  on a transmissometer with a dynamic zero
   and span. As this is a single  pass instrument on an operating stack, this was
   not possible. After the preceding standardization procedure the test was
   conducted with the stack in operation. This is not proscribed procedure.  It
   will only give relative results.  If  the in-stack opacity is reasonably stable
   an estimate of linearity is possible but it is not accurate. Linearity was
   tested by insertion of National Bureau Of Standards traceable neutral density
   filters into the light path. To assure that all of the light entering the
   photocell was attenuated, the  filter was attached to the face of the
   photocell. Three filters were  used with nominal opacities of 20,50, and 80%.
   To account for the variation in response due to stack opacity, multiple
   insertions of each filter were conducted.

C. INFRARED RESPONSE

   Infrared response was audited  by  placing an infrared filter over the photocell
   in the system. This filter passes less than 3% of visible light.  Thus the
   photocell output should be near zero or less than 5% of full output to
   indicate no infrared response.

D. ANGLE OF PROJECTION AND VIEW

   Due to the inability to bench  test the actual angle of projection and view, no
   actual figures are reported.   However, extensive experience with similar
   systems and a careful visual inspection of the installation did give some
   information.

E. RESPONSE TIME

   Response time was checked by timing the full scale response of the recorder
   while alternating turning the  source on and off.  Both upscale and downscale
   response were measured.

-------
         IV. . AUDIT RESULTS
         A. CALIBRATION

            The calibration could not be accurately determined on the instrument
            at the time of the audit.  Instead a; relative calibration was
            performed without the actual zero value.

B. LINEARITY

   The following table was developed from equations derived from the following
   relationship.  It allows the establishment of a baseline within approximately
   10% opacity.  The average opacity can then be used in the same equation to
   estimate the response of thE meter to filter  insertion with the baseline
   removed.
                                    100-T

                                      100
                 F     Filter value in % opacity
                 E     Estimated opacity of  stack in %
                 T     Total measured opacity by meter in %
WHERE
                                   TABLE,1

                 ESTIMATION OF OPACITY FROM FILTER AND METER DATA
               Data   Opacity    Opacity    Trans.
               point  observed    filter  observed
                                            Trans.   Trans.  Opacity
                                            filter  estimate  estimate
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
44
74
34
52
38
58
42
60
100
30
60
40
28
80
80
60
44
44
71
21
44
21
44
21
44
100
0
44
21
0
71
71
44
21
56
26
66
48
62
42
58
40
0
70
40
60
72
20
20
40
56
56
30
79
56
79
56
79
56
0
100
56
79
100
30
30
56
79
100
88
83
85
78
75
73
71
0
70
71
76
72
68
68
71
71
         * Point  taken  prior  to  system stability
         ** Total light path  blocked
                                                                       0 *
                                                                       12
                                                                       17
                                                                       15
                                                                       22
                                                                       25
                                                                       27
                                                                       29
                                                                      100 **
                                                                       30
                                                                       29
                                                                       24
                                                                       28
                                                                       32
                                                                       32
                                                                       29
                                                                       29

-------
   Table 2 was developed from the data on table 1. It utilizes the average
   opacity of the stack as the baseline ( E )  in the equation to determine the
   effect of the filter and the linearity of the system.
                                         TABLE 2

                     ESTIMATE OF METER RESPONSE TO FILTER INSERTION
                DATA    OPACITY   TRANS    OPACITY
                POINT  OBSERVED  OBSERVED   BASE
                    1
                    2
                    3
                    4
                    5
                    6
                   - 7
                    8
                    9
                   10
                   11
                   12
                   13
                   14
                   15
74
34
52
38
58
42
60
30
60
40
28
SO
80
60
44
26
66
48
62
42
58
40
70
40
60
72'
20
20
40
56
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
75
75
75
75
75
75
75
75
75
75
75
75
75 .
75
75
                         TRANS  ESTIMATE   FILTER
                         BASE  RESPONSE    VALUE
65       71
12       21
36       44
17       21
44       44
23       21
47       44
 7        0 *
47       44
20       21
 4        0 *
73       71
73       71
47       44
25       21
            * No filter inserted

Charts 1 and 2 illustrate the approach that was used. In chart 1 the actual
response of the meter is plotted against the known filter value.  The meter is
set up with an upscale zero based upon visible emissions observations.  Scatter
is due to changing opacities in the stack. In spite of the scatter, the points
indicate that the transmissometer is linear.  In chart 2 the average of the
estimated opacity values in table 1 (25) is used to calculate the estimated
response with the upscale value calculated out. Once again the plot is linear and
this time goes thru zero.  This demonstration of the technique was based upon a
quick evaluation of the stack opacity and is offered for demonstration purposes
only.

-------
CHART 1



tt
111


_!
E








a
in
|_
_l
E






80
72
64
56
48

40
32 ,
24 '
16
8
0
RESPONSE
D
a
a
n

" n a
g
i
-
-
i i i i \ § i i i













0 50 100
CHART 2

100
90
80
70
€0
50
40
30

20
RESPONSE
.
-
D-
-
a
-
a
UJ . _
n
n
•MM
D
10 L D -j
. 0 1
1 1 1 1 1 1 II 1 1
i












.0 50 ; 100

-------
C. INFRARED AND PHOTOPIC RESPONSE SURVEY

   The selenium detector used in the system is near photopic.  That is it matches
   the photopic response curve of the human eye on the long wave end but not on
   the ultraviolet end.  The infrared response test indicated that 5% or less of
   the response was due to infrared radiation. Considering the rough setup this
   is within acceptable limits.

D. ANGLE OF PROJECTION AND VIEW

   The angle of projection and view were not accurately determined.  However, the
   spotlight used as a source has a wide angle of projection and a more intense
   beam with a narrower angle.  The sensitivity of the system to movement of the
   light indicates that the narrovr beam has the greatest effect on the
   transraissometer.  The photocell has a wide angle of view. There are no
   aperture plates to restrict the view of the photocell and prevent it from
   being negatively biased by scattered light.

E. RESPONSE TIME

   The response time of the transraissometer is under 1 second for full scale
   response from either zero to 100 or 100 to zero.  This results in a difficult
   to read strip chart. A response time of approximately 4 seconds would be more
   useful in this system.

F. SUNLIGHT INTERFERENCE

   No appreciable interference from sunlight or ambient light was detected.

V. RECOMMENDATIONS FOR CALIBRATION/OPERATION OF PRESENT SYSTEM

   Purchase 4 inch neutral density filters of nominal opacities of 20,50,and 80%.
   This is the equivalent of transmissions of 80,50,and 20%. For the purpose of
   calibrating this system gelatin filters available from a camera store would be
   adequate. Higher quality filters would be xvasted.


PROCEDURE


   1. Establish the span by use of a certified smoke reader and adjust the meter
   to that value by adjusting the sensitivity or photocell output.

   2. Establish the transmission baseline by turning off the light source or
   blocking it.  Adjust the recorder zero to just on scale.(a live zero)

   3. Repeat steps 1 and 2 several times until consistent results are
   achieved.(no adjustment necessary)

   4. With system on,  insert the filters one at a time in the opening between the
   photocell and the stack, and record the results.  Multiple insertions are
   highly recommended.

-------
   5. Perform a linear regression of the results but do not force it thru zero.

   6. Alternately generate a calibration curve by compensating for the instack
   opacity by using equation #1. Results illustrated in table 2.


VI. RECOMMENDATIONS BOR REPAIR/MODIFICATION OF PRESENT SYSTEM
A. LIGHT SOURCE AND CONTROL SYSTEM

  ••The light should be operated at its rated voltage to assure proper light
   wavelength emissions.  This can be accomplished by installing a constant
   voltage (regulated) power supply for the light and checking that the light
   Voltage is correct with a meter at the light with the light on. Selection of
   the proper light and  its associated voltage must be in conjunction with the
   proper output  of the  photocell.

B. PHOTOCELL

   The photocell  should  be modified by the addition of an 102 filter available
   thru any retail camera store.  A better solution would be to  purchase  a
   photocell with encapsulated filter. The photocell must be producing at least
   25 millivolts  with  the selected light source to be linear. An aperture plate
   should be installed  in the photocell housing to restrict the  angle of  view to
   at least 15  degrees  preferably 10  degrees.  This would result in the detection
   of less scattered  light.

C. STRIP CHART  RECORDER

   The  strip chart recorder  does not  need  replacement.   However  it should be
   'modified by  installation  of  a 200  mfd  25  volt  electrolytic  capacitor in
   parallel with the input  to the recorder.   Additionally the  recorder should be
   set  up  to measure transmission of  light not opacity.  Thus  zero  input would
   equal  100%  opacity and 20 millivolts  input would  equal 0%  opacity.  This may
   require the adjustment of the alarm system but should cause little  difficulty.
            cell output
            millivolts
                20
                15
                10
                 5
                 0
transmission   opacity
    100%
     35%
     50%
     25%
      0%
  0%
 25%
 50%
 75%
100%
 VII RECOMMENDATION FOR REPLACEMENT OF PRESENT SYSTEM

    It is recommended that when the present unit is replaced, that the replacement
    transmissometer be a double pass instrument from one of the several major
    manufacturers.  Additionally the purchase should include a full EPA
    performance test with a condition that payment be conditional to demonstrated
    compliance with all EPA requirements.

-------
VISIBLE EMISSION OBSERVATION FORM 2

-------

-------
                                   TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing)
1. REPORT NO.
   EPA 340/1-87-002
                                                           3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
   Municipal  Waste Combustor Systems
   Operation  and  Maintenance Study
5. REPORT DATE
June  1987
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
    Roger  D.  Allen
                                                           8. PERFORMING ORGANIZATION REPORT NO.
3. PERFORMING ORGANIZATION NAME AND ADDRESS
    Allen  Consulting and Engineering
    317 Howl and Avenue
    Gary,  NC 27513
                                                           10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
 7W-6610-NASX
12. SPONSORING AGENCY NAME AND ADDRESS
    U.S.  Environmental Protection Agency
    Stationary  Source Compliance Division,  OAQPS
    401 M.  Street  SW
    Washington, DC 20460
13. TYPE OF REPORT AND PERIOD COVERED
     Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
    SSCD  Project Officer: Pamela Saunders,  EN-341  (202) 382-2889
16. ABSTRACT

        This  study was undertaken to  determine the significant operation  and
  maintenance (0 & M) considerations that  bear directly on the day  in/day out
  air  pollution compliance status of municipal waste combustors  (MWC's).   Two
  independent tasks were performed to  compile and document these 0  &  M considerations.

        First, seven MWC sites were visited to interview equipment operators and
  management personal to determine the nature of the problems they  routinely
  encounter  in operating MWC systems.   The information gathered  from  their visits
  is  reported herein through a description of 0 & M problems and solutions for
  each major subsystem in the MWC process.  The subsystems discussed  include: fuel
  preparation and handling; furnace  operation; ash handling; air pollution control
  equipment; fans, ducts and stacks; and procedural considerations  such  as startup/
  shutdown,  equipment inspection, and  recordkeeping.

        Second, a field study was conducted to determine the effect  of operating
  variables  on visible emissions (opacity) at one facility.  Ninety days  of hourly
  readings of process and opacity data were collected by the MWC staff.   An attempt
  was  then made to statistically correlate the process and opacity  data.	
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS  C. COS AT I Field/Group
    Air Pollution Emissions
    Municipal  Waste Incinerators
    Control  Equipment
    Operation  and Maintenance
    Visible Emissions
18. DISTRIBUTION STATEMENT
       Release unlimited
                                              19. SECURITY CLASS (This Report)
                                                                         21. NO. Oi- PAGES
                                              2O. SECURITY CLASS (Thispage)
                                                                         22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION is OBSOLETE

-------

-------