EVALUATION OF CONTINUOUS COMPLIANCE MONITORING REQUIREMENTS
FOR VOC ADD-ON CONTROL EQUIPMENT
EPA Contract No. 68-02-4464
Task Assignment 60
E060-1
Prepared for:
U.S. Environmental Protection Agency
Stationary Source Compliance Division
Washington, DC 20460
September, 1989
Prepared by:
Pacific Environmental Services, Incorporated
3708 Mayfair Street, Suite 202
Durham, NC 27707
(919) 493-3536
-------
CONTENTS
Page
EXECUTIVE SUMMARY iv
1.0 INTRODUCTION 1-1
2.0 VOC CEM REQUIREMENTS AND OPERATIONAL ASPECTS 2-1
2.1 NSPS/NESHAP Requirements 2-1
2.2 Continuous System Monitoring 2-5
2.3 Critical Operational Aspects 2-12
2.4 References 2-18
3.0 INSPECTION OF VOC ADD-ON CONTROL SYSTEMS 3-1
3.1 Capture Systems 3-1
3.2 Control Devices 3-11
3.3 Inspections with Portable VOC Analyzers 3-29
3.4 References 3-32
4.0 CONCLUSIONS AND RECOMMENDATIONS 4-1
4.1 Guidelines for Application of Continuous
Monitors 4-1
4.2 Continuous Monitor Performance Guidelines 4-5
4.3 Recommendations and Conclusions 4-6
4.4 References 4-8
APPENDIX A Compliance Inspection Checklists A-l
ii
-------
FIGURES
Figure Page
3.1 Typical Capture System Layout 3-3
3.2 Flowchart of Typical Carbon Adsorption System 3-20
TABLES
Table Pa§e
2.1 Applicable NSPS and NESHAP Subparts 2-2
2.2 Maximum Operating Temperatures for Some Common
Sheath Materials 2-11
2.3 PSAPCA Performance Specifications for NDIRs 2-17
2.4 NDIR Test Results 2-17
3.1 Effects of Problems with Hoods 3-7
3.2 Effects of Problems with Ductwork 3-9
3.3 Effects of Problems with Fans 3-10
3.4 VOC Concentrations Corresponding to
25 percent of the LEL 3-12
3.5 Effects of Problems with Incinerator Units 3-16
3.6 Effects of Problems with Adsorption Units 3-22
3.7 Effects of Problems with Absorption Units 3-28
m
-------
EXECUTIVE SUMMARY
The purpose of this document is to present information on the
current use of continuous emission monitors (CEMs) as compliance tools
for emissions of volatile organic compounds (VOCs). Presently, three
types of VOC monitors are in use: flame ionization detectors (FIDs);
photoionization detectors (PIDs); and non-dispersive infrared detectors
(NDIRs). This document also discusses monitoring of process related
parameters (e.g. temperature, pressure), and evaluation of VOC add-on
control systems.
This report uses terms such as continuous monitoring systems,
continuous emission monitoring (CEM), and continuous compliance
monitoring (CCM). For clarification, the term continuous monitoring
system refers to continuous monitoring of a process condition such as
gas stream temperature or pressure. The term VOC CEM is used to denote
a VOC detection instrument such as a flame ionization detector (FID),
used as a performance indicator. The term VOC CCM defines a VOC
detection instrument used to assess regulatory compliance.
The use of VOC CEMs as a continuous compliance monitor (CCM) is
limited in scope. Reasons for this include lack of applicable
performance specifications, source complexities, and instrument
limitations. However, these instruments are widely used, in conjunction
with a VOC add-on control system, as an indirect indicator of control
system performance. Where performance specifications have been
established, and source complexity problems and instrument limitations
can be successfully addressed, these instruments can be reliably used
as CCMs. For example, NDIRs have been used as CCMs on gasoline bulk
loading terminals since 1988.
RECOMMENDATIONS AND CONCLUSIONS
EPA should develop performance specifications for VOC CCMs
based on those detailed in Performance Specification 2, 40
CFR 60, Appendix B; currently no performance specifications
for VOC CEMs/CCMs exist. In addition, the NSPS regulations
may need to be slightly revised in order to implement VOC
CEMs as a compliance tool. VOC emission limitations given
in NSPS regulations typically are stated in mass emission
rates of VOC (e.g. Ib VOC/hr). Other VOC emission
limitations given in NSPS regulations specify a control
efficiency for add-on VOC control devices. VOC CEM/CCM
instruments directly measure VOC concentration, not mass
emission rates. Thus, NSPS regulations limiting mass
iv
-------
emission rates may require revision if VOC CCMs are to be
implemented.
EPA should develop a methodology to ensure relative accuracy
to ± 2 percent for solvent recovery monitoring instruments
specified in NSPS Subpart RR.
Some State and local agencies have developed VOC CEM
specifications based upon Performance Specification 2
(PS 2), as detailed in 40 CFR 60, Appendix B.
Flame ionization detectors (FIDs), photoionization detectors
(PIDs), and non-dispersive infrared detectors (NDIRs) may be
used as a CCM in some instances, but instrument limitations
and process complexities limit their use on a broad scale.
Sources operating VOC CCMs should forward a copy of vendor
maintenance requirements to regulatory agencies, who, in
turn, should ensure that the sources follow these
requirements.
Sources having more consistent gas stream characteristics
are better candidates for use of VOC CEMs as compliance
monitors (VOC CCMs).
Where VOC CEM/CCMs are applied, careful consideration should
be given to engineering of the sample conditioning systems.
Use of a portable VOC analyzer by regulatory agencies and sources
is encouraged to assist in determinations of VOC CEM/CCM system
performance, as well as to determine the integrity of the VOC
capture system.
Continuous temperature monitors provide an accurate indica-
tion of compliance for thermal incinerators.
In addition to continuous temperature monitoring, an alter-
native inspection technique using continuous or portable VOC
analyzers would be helpful in confirming adequate VOC
destruction efficiency in catalytic incinerators.
Incineration systems (both thermal and catalytic), equipped with
heat exchangers should be required to continuously monitor
pressure drop across the exchanger to ensure that the combustion
chamber is not bypassed.
Process lines using air flow recirculation are usually required to
employ an LEL indicator. Their installation is usually driven by
insurance requirements and safety concerns.
-------
CHAPTER 1
INTRODUCTION
The U.S. Environmental Protection Agency (EPA) has established new
source performance standards (NSPS) for several surface coating and
printing operations and other VOC sources. In addition, the States have
implemented EPA guidance, given in the form of control technique
guideline (CT6) documents, into regulations for existing sources. It
was originally believed that most surface coating/printing operations
would meet these requirements by using reformulation techniques (e.g.,
low solvent or waterborne coatings). However, many sources have found
reformulation difficult and have chosen to meet these regulations with
add-on controls consisting of an emission capture system and a control
system.
VOC capture and control systems do not readily lend themselves to
a "quick and easy" compliance assessment. A mass balance of solvent
usage is often used to assess emissions. However, the accuracy of this
method is limited to the ability of the source to quantify solvent
usage. Another problem involves the difficulty in determining whether
estimated solvent losses are due to poor performance of the control
device or due to inadequate capture of the solvent vapors at the process
equipment.
These facts have stirred interest in continuous monitoring systems
which can provide an indication of the efficiency of the VOC capture and
control system. Such a continuous monitoring system, properly operated,
can provide source personnel and EPA inspectors alike with pertinent and
continuous operating information concerning the capture and control
system operation. In some cases, a system can go beyond performance
indications and serve as a continuous compliance monitor.
This report uses terms such as continuous monitoring systems,
continuous emission monitoring (CEM), and continuous compliance
monitoring (CCM). For clarification, the term continuous monitoring
system refers to continuous monitoring of a process condition such as
gas stream temperature or pressure. The term VOC CEM is used to denote
a VOC detection instrument such as a flame ionization dectector (FID),
used as a performance indicator. The term VOC CCM defines a VOC
detection instrument used to assess regulatory compliance.
Continuous monitoring systems may be required on control equipment
through NSPS regulations, NESHAP regulations, or specific permit
conditions. Typically, these regulations or conditions will focus upon
specifications for continuous monitoring systems (e.g., temperature) for
1-1
-------
thermal and catalytic incinerators, and VOC CEMs (e.g., FID) for carbon
adsorbers.
This document evaluates and summarizes information available
concerning continuous monitoring of various VOC add-on controls. The
information gathered is the result of a literature search and surveys of
EPA, State, and local agency personnel and equipment vendors. It is
apparent that actual field use of VOC CEMs/CCMs, as well as specific
guidance on their performance, is limited.
Chapter 2 of this document summarizes CCM requirements which were
identified for various VOC source categories, as well as the continuous
monitors that are available, currently in use, or could be used for
various applications. In addition, information on critical operational
aspects, calibration and maintenance requirements, and relative accuracy
is presented.
Chapter 3 provides guidance on evaluation of VOC add-on control
systems. This guidance is predicated upon various parameters known to
be critical to effective emissions control and, ultimately, compliance.
Techniques for the inspection of add-on control systems are commonly
classified in guidance documents as Level 2 or Level 3. Level 2
inspections primarily involve visual observations made by an inspector
during a walkthrough evaluation of the process and control system. The
Level 3 inspection also involves these visual observations, but is
supplemented by independent measurements. In some cases where VOC CEMs
are used onsite, inspectors have employed portable VOC analyzers to
"spot check" CEM performance. Therefore, the use of portable analyzers
as an inspection tool is also discussed.
Finally, Chapter 4 summarizes the conclusions and recommendations
of this effort. The conclusions presented are the result of an
extensive survey and focus on certain concerns and shortcomings
identified regarding use of VOC CCMs. In light of these concerns,
recommendations are then provided towards establishing guidelines for
VOC CCM use.
1-2
-------
CHAPTER 2
VOC CEM REQUIREMENTS AND OPERATIONAL ASPECTS
2.1 REQUIREMENTS OF EPA REGULATIONS
There are no specific CTG requirements concerning the use of VOC
CEMs. Several NSPS standards and one NESHAP requirement specifically
discuss VOC continuous monitoring. The NESHAP requirement (Subpart V)
discusses continuous monitoring requirements for alternative operational
or process parameters (see Section 2.1.4). The VOC emission limitation
standards contained in the NSPS are typically given in a mass emission
rate (e.g. Ib/hr). Other limitations given in NSPS regulations specify
a required control efficiency for add-on VOC control equipment. VOC CEM
instruments cannot directly measure mass emission rates. Instead, they
measure VOC concentration.
Requirements for monitoring of VOC emissions and operations are
found in the various NSPS subparts and can be divided into four areas:
1) monitoring of incineration operations, 2) monitoring of solvent
recovery operations, 3) monitoring of the capture system, and
4) monitoring of other operations. NESHAP requirements are limited to
area 4. Table 2.1 provides a listing of NSPS and NESHAP subparts that
contain some requirements for monitoring VOC emissions. The following
sections describe the nature of the monitoring requirements found in
NSPS and NESHAP subparts.
2.1.1 Monitoring of Incineration Operations8
An affected facility which uses a capture system and a thermal or
catalytic incinerator to attain compliance is required to install,
calibrate, maintain, and operate temperature measurement devices. These
temperature measurement devices are to be equipped with a recording
device so that a permanent continuous record is produced. However, the
regulations do not specify a methodology for determining temperature
measurement accuracy.
8 These monitoring requirements for thermal and catalytic incinerators
are applicable to NSPS Subparts EE, MM, RR, SS, TT, WW, BBB, FFF, QQQ, and
SSS.
2-1
-------
TABLE 2.1
APPLICABLE NSPS AND NESHAP SUBPARTS
Suboart
Description
NSPS
EE
MM
RR
SS
TT
VV
WW
BBB
FFF
QQQ
SSS
NESHAP
V
Standards of Performance for Surface Coating of Metal
Furniture.
Standards of Performance for Automobile and Light Duty Truck
Surface Coating Operations.
Standards of Performance for Pressure Sensitive Tape and
Label Coating Operations.
Standards of Performance for Industrial Surface Coating:
Large Appliances.
Standards of Performance for Metal Coil Surface Coating.
Standards of Performance for Equipment Leaks of VOC in the
Synthetic Organic Chemicals Manufacturing Industry.
Standards of Performance for the Beverage Can Surface
Coating Industry.
Standards of Performance for the Rubber Tire Manufacturing
Industry.
Standards of Performance for Flexible Vinyl and Urethane
Coating and Printing.
Standards of Performance for VOC Emissions from Petroleum
Refinery Wastewater Systems.
Standards of Performance for Magnetic Tape Coating
Facilities.
National Emission Standards for Equipment Leaks (Fugitive
Emission Sources).
2-2
-------
Thermal incineration
Where thermal incineration is used, a temperature monitoring
device, equipped with a continuous recorder, is required in the
combustion zone of the incinerator. The device is to have an accuracy
of the greater of 0.75 percent of the temperature being measured
(expressed in degrees Celsius) or ± 2.5°C whichever is greater.
The facility is required to submit a report for each 3-hour period
of process operations where the average temperature of the device is
more than 28°C (50°F) below the average temperature of the device
determined during the most recent performance test.
Catalytic incineration
Where catalytic incineration is used, temperature monitoring
devices, equipped with continuous recorders, are required immediately
before and after the catalyst bed of the incinerator. The devices are
to have an accuracy of 0.75 percent of the temperature being measured
(°C) or ± 2.5°C, whichever is greater.b
A report must be submitted summarizing each 3-hour period of
process operations during which the average temperature of the device
immediately before the catalyst bed is more than 28eC below the average
temperature of the device at this location during the most recent
performance test. Further, a report is required for all 3-hour periods
during which the average temperature difference across the catalyst bed
is less than 80 percent of the average temperature difference across the
catalyst bed during the most recent performance test.
2.1.2 Monitoring of Solvent Recovery Operations
Requirements for processes controlled by solvent recovery devices
(i.e., carbon adsorbers and condensers) can be grouped into the
following three categories of compliance monitoring: 1) measurement of
solvent recovered, 2) VOC concentration level monitoring, and
3) monitoring of exhaust gas stream temperature.
Solvent Recovery Monitoring
The NSPS standards for pressure sensitive tapes and labels surface
coating processes (Subpart RR) controlled by a solvent recovery device
are required to install, calibrate, maintain, and operate a monitoring
device for indicating the cumulative amount of solvent recovered by the
device over a calendar month period. The monitoring device is to be
accurate within ± 2.0 percent. Other requirements simply state that an
affected facility using a solvent recovery system to comply with
b While this accuracy is specified for NSPS Subparts, EE, MM, RR, SS,
TT, WW, and FFF, different levels are specified for Subparts BBB, QQQ, and
SSS. These subparts require an accuracy of one percent of the temperature
being measured (°C) or ± 0.5°C, whichever is greater.
2-3
-------
emission limits is required to install the equipment necessary to
determine the total volume of solvent recovered daily.6
VOC Concentration Monitoring
An affected facility controlled by a solvent recovery device is
required to install, calibrate, operate, and maintain a monitoring
device which continuously measures and records the VOC concentration
level of organic compounds in the exhaust gases of the control device
outlet gas stream or both the inlet and outlet gas streams. The outlet
gas stream must be monitored if the percent increase in the
concentration level of organic compounds is to be used as the basis for
reporting. The inlet and outlet gas streams would be monitored if the
control device percent efficiency is used as the basis for reporting.
For a control device with multiple units arranged in parallel (i.e.,
carbon adsorption) with individual exhaust stacks, a monitoring device
is required to operate in each exhaust stack for a minimum of one
complete adsorption cycle per day for each adsorber vessel. Where inlet
gas stream monitoring is required on such a system and, depending upon
the system configuration, the affected facility may monitor the common
carbon adsorption system inlet gas stream or each individual carbon
adsorber vessel inlet stream.
This monitoring system is to be based on a detection principle
such as infrared, photoionization, or thermal conductivity, and equipped
with a continuous recorder. This continuous monitoring system is to be
installed in a location that is representative of the VOC concentration
in the exhaust vent, at least two equivalent stack diameters from the
exhaust point, and protected from interferences due to wind, weather, or
other processes.
Exhaust Gas Temperature Monitoring
An affected facility controlled by a condensation system is
required to install, calibrate, maintain, and operate, according to the
manufacturer's specifications, a monitoring device that continuously
indicates and records the temperature of the condenser outlet gas stream
(Subpart SSS).
2.1.3 Monitoring of the Capture System
Pressure sensitive tape and label coating operations and Magnetic
tape coating facilities which achieve compliance with the use of add-on
control equipment that includes a VOC capture system are required to
submit for approval a monitoring plan for the VOC capture system. This
plan must identify the parameter to be monitored as an indicator of VOC
capture performance (e.g., the power supply to the exhaust fans or duct
flow rates) and the method for monitoring the chosen parameter. The
facility is required to install, calibrate, maintain, and operate,
c NSPS, Subparts EE, MM, RR, SS, TT, and WW.
d NSPS, Subparts BBB, FFF, QQQ, and SSS.
2-4
-------
according to the manufacturer's specifications, a monitoring device that
continuously indicates and records the value of the chosen parameter.
2.1.4 Monitoring of Other Operations
Under some NSPS standards and the NESHAP (Subpart V) an
alternative operational or process parameter may be monitored if it can
be demonstrated that another parameter will ensure that the control
device is operated in conformance with the specified emission
limitations and the control device's design specifications. A
description of the parameter(s) to be monitored is required, as well as
an explanation as to why that parameter (or parameters) was selected for
monitoring.6
2.2 CONTINUOUS SYSTEM MONITORING
2.2.1 VOC CEM Operational Aspects
At present, three different types of VOC CEMs are in use as either
compliance monitors or control equipment performance indicators
depending upon the specific application. The VOC CEMs include: 1)
flame ionization detectors (FID), 2) photoionization detectors (PID),
and 3) non-dispersive infrared detectors (NDIR).
Potential problems of VOC CEMs instruments include: 1) variations
in emission stream characteristics, and 2) instrument limitations. An
example of variation in emission stream characteristics is a coating
line, where the numerous solvents used make quantification of VOC
emissions on a continuous basis difficult. In addition, the regulation
language itself may hinder instrument employment due to difficulty in
correlating instrument readings with emission regulations. For example,
it may be difficult to correlate a VOC concentration reading into a mass
emission rate (Ib/hr).
To appreciate the difficulties arising from instrument
limitations, some understanding of the operation of each instrument and
the inherent restrictions corresponding to that operation as related to
VOC emission regulations is essential.
Flame Ionization Detectors (FIDs)
This instrument employs a hydrogen flame within a combustion
chamber to ionize all organic vapors introduced to the instrument. As
the organics are combusted, positively charged carbon ions are produced
and collected on a negatively charged electrode. The accumulating ions
produce a current on the electrode proportional to the total organic
concentration. FIDs are the most common VOC CEM used and have several
advantages associated with their operation. The instrument generally
has a quick response time (much less than 15 minutes), and a linear
response between zero and span. Typically, FIDs tend to drift around
6 NSPS, Subparts VV, BBB, and QQQ; NESHAP, Subpart V.
2-5
-------
the zero point but span is stable.6 Moreover, FIDs tend to have a
fairly precise response.
However, according to vendors, FIDs employed as continuous
compliance monitors (CCMs) have two major limitations. One limitation
is that an FID is a total organic vapor analyzer. An FID will ionize
all organic vapors present within the combustion chamber and thus cannot
differentiate between organic compounds within the emission stream. The
reading obtained from the FID is indicative of the amount of carbon
within the stream, not the organic concentration except in cases where a
single organic compound is present. Therefore, the instrument must be
calibrated with the emission stream using EPA Reference Method 18 or 25
as appropriate. Use of these EPA Reference Methods may enable a
correlation between the instrument reading and actual VOC emissions
(e.g. Ibs VOC/hr) to be obtained. However, the VOC gas stream
composition must remain constant over time to ensure the instrument is
calibrated to current stream conditions. On several sources, most
notably paint or ink coating lines, emission stream characteristics may
change radically over the course of a day, meaning any calibration
performed on the instrument is quickly rendered obsolete, and hence its
use as a continuous compliance monitor is limited. An FID may still
serve as a performance indicator for VOC control equipment present.
The second problem with FIDs concerns the response of the
instrument to various organic compounds. FIDs are often calibrated with
methane or propane, and the instrument reads organic concentration as
parts per million methane or propane. Different organic compounds will
cause the instrument to respond in extremely different ways. The ratio
between the instrument response when reading a given VOC compound and
the actual concentration of that compound is termed the response factor.
This factor is used to obtain an estimation of the stream concentration,
but it requires knowledge of stream constituents which is not usually
available on a continuous basis. Use of this factor introduces several
sources of error including the imprecision of response factor numbers6
and lack of knowledge of stream constituents on a continuous basis.
Photoionization Detectors (PIDs)
This instrument is quite similar to an FID except that a light
beam is used to ionize the organic compounds within the stream. The
ionization detector consists of a chamber containing a pair of
electrodes. A positive potential is applied to one electrode.
Subsequently, a field around the electrode is induced by the potential
and ions formed by the light beam are driven to the collecting electrode
where a current proportional to the concentration is measured.
Like FID instruments, PIDs usually exhibit a quick response time
and a linear response between zero and span. PID instruments tend to be
fairly stable around both zero and span, and have precision similar to
FID instruments.7
A significant limitation to PID instruments is their inability to
read organic compounds with high ionization potentials (i.e., aliphatic
(straight chain) hydrocarbons). PIDs cannot therefore be used as a
total hydrocarbon analyzer on streams containing straight chain
2-6
-------
hydrocarbons (e.g., methane, ethane, propane, etc.) because these
compounds will not be detected. Moreover, reliability problems with
bulbs have been reported.3
Also, PIDs share the same limitations of FIDs. PIDs are typically
employed when some degree of specificity is desired and aliphatic
hydrocarbons are not a concern. They may be useful as CCMs in selected
cases but obviously cannot be recommended as a broadly applicable
instrument.
Non-Dispersive Infrared (NDIR)
This instrument differs from FIDs and PIDs in that ionization of
organic compounds does not occur. Instead, the instrument relies upon
absorption of infrared light of a discrete wavelength to measure organic
concentration.
The typical NDIR instrument consists of two identical infrared
emitters. The beams generated from each emitter are modulated, and pass
through filters and measuring cells into an energy receiver. One
measuring cell, called the reference cell, is filled with a gas that
does not absorb infrared radiation (e.g., N2). The other cell called
the analysis cell, is filled with the gas being analyzed. The infrared
radiation passing through the analysis cell will be absorbed in an
amount directly related to the concentration of organic compounds within
the gas stream. The beams are then directed to the energy receiver
which consists of two chambers separated by a flexible non-permeable
membrane. Both chambers are filled with a mixture of argon and the
organic compounds to be measured. Absorption of the infrared radiation
heats both energy receivers causing a corresponding pressure increase
resulting in membrane deflection. The analysis receiver will have less
energy available to heat its chamber, however, because some radiation
has been absorbed in the measuring cell. The reference receiver, on the
other hand, has virtually all the radiation energy directed to it. The
membrane will, therefore, deflect towards the analysis receiver. The
amount of deflection is proportional to the concentration of organics in
the gas stream.8
As a result of the mode of operation, several potential problems
emerge when trying to employ NDIRs as a broad based continuous
compliance monitor. First, the NDIR generates infrared radiation at a
discrete wavelength. Typically, the wavelength is selected based upon
absorption characteristics of the organic compounds to be measured.
However, if the gas stream contains a mixture of organic compounds, as
is often the case in coating and printing applications, it is not
possible to select a wavelength that will result in an accurate
concentration measurement of all organic compounds on a continuous
basis. Additionally, it may be difficult to correlate a concentration
reading with the emission regulation (e.g. Ibs VOC/hr). For an NDIR,
the organic compound concentration measured is based upon the physical
deflection of the membrane, which in turn is a function of the radiation
absorbed in the analysis cell. Different organic compounds will absorb
infrared light in various amounts, and, in fact, at a given wavelength
some organic compounds may not absorb infrared radiation at all. To
2-7
-------
obtain an accurate measurement requires knowledge of the organic
constituents in the emission stream.
NDIRs are better suited for use as VOC CCMs for applications
having consistent emission stream characteristics and known organic
constituents. Under these conditions, the instrument has adequate
performance characteristics to warrant considerations as a continuous
compliance monitor. In fact, NDIRs have been successfully employed as
compliance monitors on gasoline bulk terminals for over a year. ' The
instruments typically exhibit a linear response, have very little drift
and a quick response time. Moreover, NDIRs are usually very precise,
provided the emission stream exhibits consistent characteristics over
time. One local agency has indicated that NDIRs may be broadly
applicable as a VOC CCM on numerous bulk terminals.
2.2.2 Monitoring Systems for VOC Control Techniques
The fact that VOC CEMs may in some cases be inappropriate as a
compliance monitor should not necessarily dissuade sources from using
VOC CEMs as a performance indicator. In many cases, information
provided by VOC CEM systems can be valuable in determining whether a
control system is operating correctly. For example, a source may employ
a VOC CEM to help monitor the control device performance. In some cases
however, the information provided by VOC CEMs may not be sufficient to
adequately gauge control device performance. Often, monitoring of
process parameters is also necessary to ensure control device
performance is satisfactory. These process parameters (e.g. incinerator
temperature) vary depending upon the specific control device.
The three prevalent VOC control devices used include thermal
incinerators, catalytic incinerators, and carbon adsorbers. Continuous
temperature readings are required for both incinerator types. For
thermal incinerators, combustion chamber temperature is monitored, while
the temperature before and after the catalyst bed is monitored on
catalytic incinerators. State and local agency permits often require
temperature monitors for thermal and catalytic incinerators. Monitoring
requirements for carbon adsorbers typically involve a VOC CEM (e.g. an
FID) on the bed outlet gas stream to ensure breakthrough does not occur.
State and local agencies will typically determine performance
specifications for VOC CEMs used with a carbon adsorber on a case-by-
case basis.1'3
In cases where boilers destruct recovered waste solvents from a
carbon adsorber, permits usually require continuous monitoring and
recording of combustion zone temperature, and use of a continuous flow
measurement instrument and recorder to quantify the amount of solvent
supplied to the solvent burner on the boiler.
Thermal Incinerator Monitor Requirements
Monitoring combustion temperature is sufficient for thermal
incinerator VOC destruction performance evaluations. This is based upon
the premise that combustion chamber temperature is the fundamental
indicator of VOC destruction efficiency. However, additional parameters
may need monitoring to ensure all VOCs within the emission stream reach
2-8
-------
the chamber. One coating source has reported that monitoring of
incinerator heat exchanger pressure is necessary to ensure all organic
compounds are ducted to the control device.9 Monitoring heat exchanger
pressure provides an indication of exchanger integrity. For example, if
a leak were to develop in the heat exchanger (leading to the release of
organic compounds in the gas stream), a change in pressure differential
would alert operators to check for such a leak, and effect repair.
Catalytic Incinerator Monitor Requirements
NSPS monitoring requirements for catalytic incinerators usually
involve continuous temperature readings before and after the catalyst
bed to ensure catalyst activity is sufficient. In addition, monitoring
of the pressure drop across the catalyst bed ensures an adequate amount
of catalyst is present. According to vendors, the catalyst activity
should be checked every three years, or whenever the temperature
difference across the catalyst bed is consistently beyond the normal
operating range. These requirements are sufficient to ensure the
incinerator is operating within specifications, but additional
requirements (e.g., heat exchanger pressure) may be necessary to ensure
all VOCs within the emission stream are introduced to the incinerator.
Temperature Monitoring Systems
Continuous temperature monitoring systems are very common within
industry and can exhibit a high degree of precision and accuracy. The
regulations reflect this fact, and require the monitoring system to
obtain a high degree of accuracy. The systems themselves consist of a
thermocouple(s) which generates an electrical signal. Measurement of
electrical signals can be performed with an extremely high degree of
precision, accuracy, and reliability. The thermocouple itself however,
while being simple and rugged, generally exhibits non linear performance
characteristics, imprecision, and somewhat poor accuracy when measuring
temperatures over a wide range.10 These shortcomings can be minimized
significantly if only a relatively narrow range of temperatures must be
measured. This may necessitate choosing a temperature range where
accuracy is deemed the most critical. Within a fairly narrow
temperature range however, temperature monitoring systems employing
thermocouples should exhibit performance characteristics that meet
requirements stated in NSPS regulations.
In examining a continuous temperature monitoring system, ensuring
the correct thermocouple type and sheath are employed is important.
Selection of the correct thermocouple and sheath will help prevent
damage due to high temperatures and attack by sulfurous compounds and
reducing agents if these are present. A variety of thermocouple types
are used by industry. A brief summary of some common types is provided
below:11
Type K - This is the most common type of thermocouple used for VOC
inspections due to the broad temperature range of -400 °F
to + 2300 °F. The thermoelectric elements must be
protected by a sheath since both wires are readily
attacked by sulfurous compounds and most reducing agents.
This sheath must be selected carefully to ensure that the
2-9
-------
thermocouple can withstand the maximum temperature that
the unit will be exposed to. The positive wire is nickel
with 10 percent chromium (trade name - chromel) and the
negative wire is nickel with 5 percent aluminum and
silicon (trade name - alumel).
Type E - These generate the highest voltage of any thermocouple
but are limited to a maximum temperature of 1600 °F.
Higher voltage generation is an advantage because it
allows for greater differentiation and hence accuracy in
temperature readings. The positive wire is nickel with
10 percent chromium (chromel) and the negative wire is a
copper-nickel alloy (constantan).
Type J - These have a positive wire composed of iron and a
negative wire composed of a copper-nickel alloy
(constantan). They can be used up to 1000 °F in most
atmospheres and up to 1400 °F if properly protected by a
sheath. They are subject to chemical attack in sulfurous
atmospheres.
Type T - These can be used under oxidizing and reducing
conditions. However, they have a very low temperature
limit of 700 °F. They are composed of copper positive
wire and a copper-nickel alloy (constantan) negative
wire.
Type R and S - These can be used in oxidizing or inert conditions
to 2500 "F when protected by nonmetallic protection
tubes. The Type R thermocouples are composed of a
positive wire of platinum with 13 percent rhodium and a
negative wire of platinum. The Type S thermocouples have
a positive wire of platinum with 10 percent rhodium.
Both types can be subject to calibration drifts to lower
temperature indications due to rhodium diffusion or
rhodium volatilization.
Type B - The positive wire is composed of a platinum with 30
percent rhodium and the negative wire is platinum with 6
percent rhodium. These are less sensitive to the
calibration drift problems of Type R and S thermocouples.
They can be used to a maximum temperature of 3100 "F when
protected by nonmetallic protective tubes.
The maximum temperature a thermocouple can withstand is dependant
upon wire compositions and on the sheath type around the thermocouple
junction. Table 2.2 presents maximum operating temperatures associated
with each sheath type.11
Pressure Monitoring Systems
Common static pressure gauges may be used to evaluate the pressure
drop across a heat exchanger and across a catalyst bed. Common types of
static pressure gauges employed by industry include U-tube manometer,
inclined manometers, and diaphragm gauges. The inclined manometer is
2-10
-------
TABLE 2.2
MAXIMUM OPERATING TEMPERATURES FOR
SOME COMMON SHEATH MATERIALS
Sheath Material Maximum Temperature "F
Aluminum 700
304 Stainless Steel 1,650
314 Stainless Steel 1,650
Inconel 2,100
Hastelloy 2,300
Nickel 2,300
2-11
-------
generally considered the most accurate for pressures below 10 inches of
water. U-tube manometers may be used for pressures up to about 40
inches of water. Diaphragm gauges are usually accurate to ± 5 percent
of the instrument scale.
The operation of a manometer is quite simple. The instrument
typically consists of a U-tube or inclined tube filled with a liquid. A
gauge adjacent to the tubes provides an indication of the pressure
differential between the two ends. Depending upon relative pressures to
be measured, the specific gravity of the fluid can be adjusted to
correspond with the pressure. Several different manometer types are
employed by industry. The type of manometer is dependant upon the tube
ends. An open-end manometer consists of one tube end exposed to
atmospheric pressure and the other exposed to the pressure of the fluid
to be measured. A differential manometer consists of both ends exposed
to the fluid at different points in the process line. The monitor
measures differential pressure across a given process component (e.g.
catalyst bed). A sealed end manometer consists of one enclosed end and
one open end exposed to the fluid to be measured. If the open end is
exposed to the atmosphere, the device is called a barometer.
A diaphragm gauge is composed of two chambers separated by a
flexible diaphragm. The diaphragm material and/or thickness is
typically dependant upon the pressures to be measured. Pressure is
transmitted to the chambers via ports which serve the same function as
manometer tubes. The diaphragm deflects when unequal pressure exists
between the two parts. The deflection is mechanically transmitted to a
dial which provides a pressure differential indication.
2.2.3 Monitor System Costs
A monitoring system which includes a VOC measuring device, sample
lines, and continuous monitor/recorder costs about $20.000 per system
with estimates ranging between $15,000 and $30,000.1'5fT2 Factors which
influence cost include: heated/non-heated sample lines, type and vendor
of VOC CEM employed, and continuous monitoring/recording requirements.
The Omega Temperature Handbook10 provides a comprehensive listing
of temperature measurement devices and costs. Prices typically average
$50 per ultra high temperature thermocouple probe with electrical
connections, and approximately $2,000-$4,000 for a precision continuous
monitoring/recording device, with a temperature to analog converter.
2.3 CRITICAL OPERATIONAL ASPECTS
Critical operational aspects of continuous monitors can be divided
into: 1) relative accuracy of the instruments, 2) calibration and
maintenance requirements, and 3) other critical operational aspects.
Each of these are discussed in more detail below.
2.3.1 Relative Accuracy
Relative accuracy is an indication of the instrument's ability to
correctly identify the value of the process parameter to be read (i.e.,
2-12
-------
temperature or VOC concentration), and is usually expressed on a percent
basis. For example, if an instrument indicates a VOC concentration of
95 ppm when the actual concentration is 100 ppm, the instrument is 95
percent accurate, or the accuracy is within 5 percent.
Instrument accuracy is very important in cases where NSPS
regulations specify that a methodology for VOC CEMs must be found to
verify accuracy to within ± 2 percent (Subpart RR). The majority of EPA
and State agency staff contacted have not defined such a methodology,
and many do not believe the instruments are capable of determining
instrument accuracy to within ± 2 percent. However, this should not
prevent EPA, State and local agencies from developing performance
standards and imposing VOC CCMs on sources not subject to subpart RR.
One local agency defined a methodology on gasoline bulk terminals and
has imposed NDIR analyzers as compliance monitors. Another local
agency requires FID analyzers as continuous compliance monitors even
though this agency does not believe FIDs can consistently read VOC
concentration to within + 2 percent (see Section 2.3.4).
2.3.2 Calibration and Maintenance Requirements
In general, calibration and maintenance requirements for VOC CEMs
and temperature monitors employed by source owners follow manufacturer's
guidelines and recommendations. Typically, the EPA, State or local
agency does not specify requirements beyond those of the manufacturer.
Despite possible procedural differences between manufacturers, there are
several criteria common to all calibration methods, including:
1) accuracy, 2) response time, 3) response linearization, 4) precision,
and 5) drift. Of these five criteria, response linearization is usually
not included as part of applicable calibration requirements.
Nonetheless, it is a common procedure used in industry. Manufacturers
recommend that these procedures be performed on a daily basis using
calibration gases at zero and full scale values.
Accuracy
Accuracy, usually expressed on a percent basis, provides an
indication of the instrument's error in detecting a given VOC
concentration. For example, if the actual concentration of calibration
gas is 200 ppm while the instrument indicates 190 ppm, the accuracy of
the monitor is 95 percent.
Response Time
Typical EPA guidelines for other continuous monitoring systems
(e.g SO,, NO , CO systems) require a system response time of less than
15 minutes for an instrument to be considered operating continuously.
The response time is measured as the time required for the entire sample
transport system to obtain 95 percent of the final stable value, after
an input step change.
2-13
-------
Response Linearization
This test provides an indication of the instrument deviation from
a response predicted by a straight line between zero and span of the
instrument.
Precision
Precision, also called repeatability, is conducted to measure the
instrument's ability to yield the same result for repeated analysis of
the same sample.
rift
Drift provides an indication of the instruments' ability to
continuously monitor a constant concentration without significant
deviation over time. Typically, this test occurs over some
predetermined time interval (e.g., 2 hours), while a constant VOC
concentration is introduced to the instrument. Drift checks may be
performed less often than the other four calibration checks due to the
time involved.
2.3.3 Other Critical Operational Aspects
FIDs, PIDs, NDIRs and temperature monitors have different
operational aspects which are critical to maintaining satisfactory
performance.
All FID instruments require fuel and combustion air to maintain
flame stability. Both fuel (hydrogen) and combustion air are usually
controlled by pressure regulators to ensure adequate and constant flow
to the combustion chamber. If constant pressure on the fuel or
combustion air is not maintained, flame instability will result and
system performance will suffer. Another critical operational aspect
concerns the level of moisture present in the sample line. Care must be
taken to ensure sample moisture does not degrade flame stability and
promote combustion chamber corrosion or performance will suffer. Often,
the sample line is heated to decrease the relative humidity and prevent
moisture related problems.
PID instruments are inherently simpler than FID units because
hydrogen fuel and combustion air are not necessary for operation.
However, problems with bulb reliability have been noted3, causing
incorrect readings. Moreover, different organic compounds require
different levels of energy to ionize the compounds. Bulb ionization
energy must be greater than the ionization energies of the organic
compounds to be monitored. If bulb energy is too weak, the organic
compounds will not ionize and false readings will result.
NDIR instruments differ fundamentally from either FIDs or PIDs in
that absorption, rather than ionization is used to read organic compound
concentration. These instruments use infrared radiation applied at a
single wavelength to determine concentration. The radiation can be
generated at a desired wavelength by passing light through a lens filter
or series of filters. Dirt and moisture accumulation on the filter(s)
2-14
-------
influences the energy and wavelength of the radiation, adversely
affecting results. Often, the sample line must be filtered and heated
to prevent these problems from occurring. Likewise, incorrect alignment
of the filter will affect results. NDIR instruments are, therefore,
usually best suited for immobile applications and where dirt and
moisture are less likely to become entrained in the sample line.
The most frequently mentioned aspect to ensure satisfactory
performance of continuous temperature monitoring systems involves
correct placement of the thermocouples including use of shields to
prevent effluent "backwash", which may cause readings lower than the
actual value. Additional considerations include use of the correct
thermocouple type, and frequent calibration of the devices.
2.3.4 State/Local Operational Requirements
Most VOC CEMs in use as compliance monitors (as opposed to
performance indicators) are installed in response to State or local
agency performance specifications. No EPA promulgated performance
specifications for certification of VOC CCMs exist. Some States and
local agencies therefore have selected certification tests based upon
appropriate procedures such as 40 CFR 60, Appendix B, Performance
Specification 2.1'3
Specifically, two local agency requirements are discussed below.
The Pudget Sound Air Pollution Control Agency (PSAPCA) successfully
implemented NDIRs on gasoline bulk terminals in 1988. According to
PSAPCA personnel, several local agencies in Caifornia are now also
requiring NDIRs on bulk terminals. The Texas Air Control Board (TACB)
is requiring VOC CCMs on several sources, although no such system is yet
operational.
Puoet Sound Air Pollution Control Agency (PSAPCA)
The PSAPCA requires hydrocarbon monitors (NDIRs) to measure
concentration of gasoline at the outlet of the vapor recovery units
(VRU) on new bulk loading terminals. The bulk loading terminals NSPS
limits VOC emissions to 35 mg VOC/1 gasoline loaded. The terminals are
required to periodically furnish the PSAPCA with strip chart records to
verify that the VRUs do not exceed an outlet hydrocarbon level of 2
percent propane. PSAPCA has imposed this concentration limit (2 percent
propane) through specific permit conditions as a surrogate to the NSPS
regulation of 35 mg VOC/1 gasoline loaded. According to PSAPCA, the 2
percent propane limit is based upon source test data taken of gasoline
loading terminals, and corresponds to the NSPS requirement of 35 mg
VOC/1 gasoline loaded. This concentration value is an enforceable limit
of the permit, with the NDIR instruments determining compliance with
permit conditions. In fact, the source is required to take action when
emissions exceed 1 percent propane, and shut the process down if
emissions exceed 2 percent propane.
The determination of a VOC outlet compliance limit involves
parameters that may vary from terminal to terminal. In addition to VOC
concentration, vent gas flow, vent temperature, vent pressure, and total
2-15
-------
product loaded are necessary to determine compliance with the NSPS
regulation.5
PSAPCA has selected appropriate performance specifications for the
NDIR monitors based upon procedures detailed in 40 CFR 60, Appendix B,
Performance Specification 2, and Appendix A, Method 25B. Table 2.3
summarizes the standards PSAPCA has imposed upon VOC CEMs for VRUs at
bulk loading terminals.
The specifications outlined in Table 2.3 were used to test two
recently installed VOC CEMs made by different manufactures at VRUs on
bulk terminals. The tests were conducted over seven days to determine
performance under typical operating conditions. Table 2.4 details the
test results. All numbers are presented as percent difference. Percent
difference is calculated as follows: Assume a calibration gas has an
actual concentration of 500 ppm, and the instrument reading is 495 ppm.
The percent difference is the absolute value of the instrument reading
minus the actual concentration (i.e., 5), divided by the actual
concentration, and multiplied by 100. For the example case, the percent
difference equals 1 percent.
The test results show NDIR analyzers perform satisfactorily to
meet the applicable standards. The NDIR analyzers have performed
extremely well during unannounced instrument inspections to test
calibration accuracy and precision by PSAPCA. PSAPCA personnel are
confident in the ability of the NDIR analyzers to monitor compliance
with the applicable NSPS regulation. The NDIRs have been in use for
over a year to date with minimal operational problems.
Texas Air Control Board (TACB).
The TACB is requiring on a case-by-case basis, continuous
monitoring of VOC emissions.3 In some cases, the VOC CEMs are used as
performance indicators and in others they are used to determine
compliance.
At present, VOC CCMs have been installed on several sources but
are not yet operational. TACB personnel have stated that monitor
certification procedures will typically employ methods 25A or 25B, and
Method 18 (if necessary). Performance specifications for monitoring
systems are based on procedures detailed in Performance Specification 2
as stated in 40 CFR 60, Appendix B.
2-16
-------
TABLE 2.3
PSAPCA PERFORMANCE SPECIFICATIONS FOR NDIRs
Parameter
Specification
Accuracy8
Calibration Error8
Zero Drift-24 hour
Calibration Drift-24 hour
Response Time
Operational Period
< 20% of the mean value of the
reference method test data, or 0.5%
propane, whichever is greater.
< 20% of 50% calibration gas, or 0.5%
propane, whichever is greater. < 20% of
90% calibration gas or 0.5% propane,
whichever is greater.
No specification
2.5% of span
15 minute maximum
168 hour maximum
8 Expressed as the absolute mean plus 97.5 percent confidence interval
of a series of tests.
TABLE 2.4
NDIR TEST RESULTS
Parameter8
Accuracy, percent
Mid-Range Calibration Error,
percent
High-Range Calibration Error,
percent
Zero Drift, percent
Calibration Drift, percent
Response Time, seconds
Brand A
0.11
0.09
0.01
0.03
0.05
<20
Brand B
0.12
0.05
0.15
0.03
0.50
<20
Recommended
Ranqe
0.50
0.50
0.50
N/A
0.50
<15 minutes
All values given in percent difference.
2-17
-------
2.4 REFERENCES
1. Telecon. Sink, M. Pacific Environmental Services, Inc., with
Jim Nolan, Pudget Sound Air Agency. June 30, 1989.
2. Telecon. Sink, M. Pacific Environmental Services, Inc., with
Kris Hansen, AM Test Inc. July 10, 1989.
3. Telecon. Sink, M. Pacific Environmental Services, Inc., with
Bill Chaffin, Texas Air Control Board. July 10, 1989.
4. U.S. Environmental Protection Agency. Gaseous Continuous
Emission Monitoring Systems-Performance Specification Guidelines
for S02, NOX, C02, 02, and TRS. EPA-450/3-82-026. October 1982.
5. John Zink Company, Letter with attachment to Pacific
Environmental Services, Inc., June 28, 1989.
6. Telecon. Sink, M. Pacific Environmental Services, Inc., with
Russell Rever, Ratfisch Instruments. July 11, 1989.
7. U.S. Environmental Protection Agency. Study of Benzene
Continuous Emission Monitoring Systems for the Gasoline Bulk
Storage Industry. Contract No. 68-02-3887. May 1987.
8. Pollution Engineering. Continuous Stack Gas Monitoring. Part
One: Analyzers. Philip Wolf. June 1975.
9. Telecon. Sink, M. Pacific Environmental Services, Inc., with
Dave Ellison. Alusuisse Corp. June 28. 1989.
10. Omega Temperature and Pressure Manuals, Vol. 26, 1988.
11. U.S. Environmental Protection Agency. APTI Field Inspection
Notebook. Draft Report. August 1987.
12. Conversation. Osbourn, S. Pacific Environmental Services, Inc.,
with Paul Reinermann, EPA Region IV. June 29, 1989.
2-18
-------
CHAPTER 3
INSPECTION OF VOC ADD-ON CONTROL SYSTEMS
An add-on emission control system consists of an emission capture
system followed by an emission control device. In this chapter, the
capture system is discussed first, followed by a discussion of control
devices. If either one of these two components of the control system is
operating less effectively than originally designed, pollutant emissions
will increase.
The capture system collects VOCs at their point of release in the
process and delivers them to a control device, which removes them from
the air stream. The efficiency of the overall control system is
directly related to the efficiency of both the capture system and the
control device. For example, if the capture system collects 90 percent
of the VOC released by a coating process and delivers the VOC to a
control device that removes or destroys 90 percent of the VOC, then the
overall control system VOC removal efficiency is 81 percent (90 percent
x 90 percent = 81 percent).
Inspection of a plant's VOC capture and control system should
start with the point of emission and proceed toward the emission control
device. The diversity of sources, capture systems, and VOC control
devices makes it impossible to describe a "typical" facility that will
be encountered in the field. Therefore, it is necessary to individually
assess each situation and diagnose the specific combination of VOC
source, capture system, and control device.
During compliance inspections, it is often difficult to state with
certainty that the capture system and control device(s) are achieving
the efficiencies required. For example, it may be difficult to confirm
that a 90 percent capture efficiency, required by permit conditions, is
being attained. In many cases, the determination of facility compliance
is strictly an engineering judgement on the part of the inspector.
Therefore, while the following discussion presents inspection procedures
and checklists, also included is general background information
regarding process equipment and operational considerations. Taken as a
whole, this guidance can be used by an inspector as the basis for an
engineering judgement as to the degree of compliance being attained.
3.1 CAPTURE SYSTEMS
It should be noted that this document attempts to consider VOC
sources identified as being subject to requirements for continuous
monitoring of their emissions and operations. These sources were
previously discussed and listed in Table 2.1. The majority of these
3-1
-------
sources are surface coating and printing operations. Therefore, the
discussion of capture systems is directed toward configurations typical
of these source categories. Instances will be noted where inspection
procedures for a source type, such as bulk gasoline terminals, may
differ from typical surface coating/printing type operations.
The capture system typically consists of three main parts: hoods
that trap the emissions, ductwork that transports emissions to the
control device, and a fan that supplies the energy necessary to move the
emissions through both the capture and control systems. These elements
are graphically displayed in Figure 3.1.1 It should be noted that the
fan may be located either upstream or downstream of the control device.
In order to achieve the overall percent emission reductions
desired at most VOC emission sources, capture system efficiencies must
typically be in the range of 90 percent or greater. During compliance
inspections, it is often difficult to state with certainty that the
required capture efficiencies are being achieved. Actual source testing
of the capture system provides the most accurate estimate of capture
efficiency.
In order to effectively capture VOC emissions, a source may
install a total enclosure. If the design of this total enclosure meets
certain EPA guidelines, indicating its efficiency is essentially 100
percent, then no performance testing of capture efficiency would be
necessary. The determination of capture efficiency is relatively
straightforward where an enclosure surrounds the emission source and all
air from it is directed to the control device. Similarly, if the source
is contained in a room or building whose entire ventilation air is
directed to the control device, capture efficiency is essentially 100
percent. These are examples of total enclosures. The only openings to
a total enclosure are those which allow raw materials to enter and
product to exit the process, the exhaust duct, and any requisite inlet
air openings. A "total enclosure" should meet the guidelines outlined
in the following paragraphs.
Based on some quantitative determination of the average air
velocity through the inlet air or makeup air openings in the enclosure,
one can conclude that its efficiency can (or cannot) be presumed to be
100 percent. The first requirement for total capture is that the
enclosure function at lower than atmospheric pressure so that air flow
is inward at all openings. This condition presupposes that any forced
makeup air is introduced to the enclosure at a rate that is less than
the rate of evacuation. The second requirement is that the average
velocity through all openings be at least 60 meters per minute (m/min)
(200 feet per minute (fpm)). The inspector may use a velometer to
determine air velocity at the openings. A traverse should be made
across the openings and the readings averaged in a consistent manner.
This value could also be calculated by dividing the total net volumetric
discharge from the enclosure (cubic meters per minute), if known, by the
total cross-sectional area of all openings (square meters).
Even if the pressure inside an enclosure was shown to be negative
and the average face velocity through all openings was calculated to be
60 m/min (200 fpm), large openings, which might be necessary for the
3-2
-------
CO
CO
fill]
ENCLOSURE
RECEIVING HOOD
CAPTURE HOOD
HOODS
FAN
DUCTS
DUCTWORK FAN
INCINERATOR
ADSORBER
ABSORBER
I
CONDENSER
CONTROL DEVICE
AMBIENT
AIR
DISCHARGE
Figure 3.1 Typical collection hood/ductwork/fan/control device layout.
-------
passage of raw materials, could allow momentary eddies and backdrafts.
For this reason, any source of VOC emissions inside the enclosure should
be at least one equivalent opening diameter from each opening.
Finally, if an enclosure opening is located in close proximity to
an exhaust hood or duct, air that is pulled into the enclosure through
the opening may be channeled directly into the hood or duct, resulting
in a short-circuiting of the system. In this case, the face velocity at
this opening is probably much greater than at others. To prevent this
short-circuiting of the system, all openings should be a minimum of
three equivalent hood or duct diameters from any hood or duct through
which the enclosure is evacuated.
If one or more of the above requirements are not satisfied, the
source will have to increase the evacuation rate of air from the
enclosure, make appropriate design changes, or consider the enclosure to
be partial. If the capture system cannot be classified as a total
enclosure, then the capture system inspection procedures detailed in the
following paragraphs (Sections 3.1.1 through 3.1.3) can be used as the
basis for an engineering judgment as to the degree of capture system
efficiency being attained. If such a determination is still not
possible, then compliance testing would be required to determine the
actual efficiency being attained.
The following discussion reviews the basic principles of
ventilation systems without getting into specific design information.
Inspections of capture, as well as control systems should be conducted
with the assumption that each was properly designed for its specific
application. The following guidance is directed at determining whether
the system works as it was originally designed, accounting for equipment
maintenance, operating conditions, and whether the system has been
significantly modified such that the original design is now
insufficient. Regardless of how well the system was originally
designed, the purpose of a compliance inspection is to determine if the
capture and control system is properly operating now.
3.1.1 Hoods
The hood is an integral part of the capture system on a surface
coating or printing operation. If VOCs are not initially captured as
they are emitted from the surface coating applicator, flash-off area,
curing oven, or other emission point in the process, the entire capture
and control system will be ineffective in limiting VOCs.
A hood is generally defined as any point where air is drawn into
the ventilation system to capture or control emissions. There are three
major types of hoods, each working on a different principle (see Figure
3.1).1
1. Enclosures are hoods that surround the emission sources as
much as possible. Emissions are kept inside the enclosure by
air flowing in through openings in the enclosure. The more
complete the enclosure, the less airflow is needed for
control. During compliance inspections, it is often
difficult to state with certainty that required capture
3-4
-------
efficiencies are being achieved. However, if it can be
demonstrated that the capture system meets certain
guidelines, it can be classified as a "total enclosure", and
its efficiency can be presumed to be 100 percent. Guidelines
for defining a total enclosure were previously presented on
pages 3-2 through 3-4.
2. Receiving hoods are designed and positioned to catch emission
effluents as they are released by the process. The hood is
located in such a fashion as to take advantage of the natural
flow of VOCs. These hoods allow workers to operate in and
around the process; however, their effectiveness may be
reduced by cross drafts in the workroom.
3. Capturing hoods make full use of capture velocity to draw in
emissions from the process. This hood is widely used, since
it can be placed alongside the emission source rather than
surrounding it as with an enclosure. The primary disadvan-
tage is that large air volumes may be needed to generate an
adequate capture velocity at the emission source. A second
disadvantage is that the "reach" of most capturing hoods is
limited to about 2 feet from the hood opening. This reach
will decrease if the system air flow decreases.
Any inspection of a surface coating or printing facility will
probably reveal one or more of the three major types of hoods. The
problem that confronts the inspector is to ascertain if the hood is
working properly. The following guidance can be used by an inspector as
the basis for an engineering judgment as to the proper functioning of
the capture system.
VOC emissions can often be visually observed in areas of high
concentration. They have the appearance of "wavy lines" when viewed
against certain backgrounds. This phenomenon can also be observed with
spilled gasoline evaporating from a service station in the summer.
These emissions would be noticeable at the entrance to an enclosure-
type hood, on the perimeter of a receiving hood, or on the side of the
emission source that is furthest from a capture hood.1
Escaping VOCs might also cause a distinct solvent odor in the room
where the hood is located. The existence of this odor depends on the
type of VOC and its concentration. Yet, if it is present, it can signal
a possible capture system problem.
Touch can also help diagnose a malfunction. While capture
velocities for VOC emissions are usually very low, placing your hand
near the capture hood can sometimes help detect if there is a noticeable
air flow toward the hood, indicating that sufficient draft exists in the
system. Conversely, if you feel an air velocity going up and around the
hood, it is a good indication that there are escaping VOCs.
Instruments, such as a velometer, are available that will assist
in determining if there is an air flow into a hood, or if the air (and
VOC) is bypassing the hood and going into the plant. With the
velometer, the velocity of the air flowing into the hood and capture
3-5
-------
system can be measured. This data can be checked against the original
design and used as a comparison to data obtained during the next
inspection of this plant.1
Another instrument that might prove valuable in inspecting capture
hood efficiencies is the smoke tube. This device emits a smoke trail
which when placed near a hood will follow the flow of air into or around
the hood. Smoke tubes can only provide a qualitative estimate of
airflow. Nonetheless, this estimate can prove very useful in
determining if VOC effluent is flowing into a hood.
Other visual clues to potential capture system problems include
clogged hood filters and noticeable holes or tears which would reduce
the effectiveness of the hood operation. Further, if original process
equipment has been significantly modified without a concurrent change in
the hood or capture system, the original design should be re-evaluated.
A summary of some problems that may arise with hoods, and the
effects that these problems would have on operations or emissions can be
found in Table 3.11. A checklist summarizing items to be checked during
a hood inspection is presented in Appendix A.
The capture of VOCs at bulk gasoline terminals differs somewhat
from that for surface coating and printing operations. Rather than
capture hoods, terminals maintain vapor-tight connections at loading
arms, fittings, and vapor lines. Specifically, no vapor leaks >100
percent of the lower explosive limit (LEL) are permitted during loading.
In addition, backpressure measured at the point of vapor collection
cannot exceed 18 inches water gauge. In order to determine compliance
with these requirements, measurements are required of the percent LEL
and backpressure during gasoline loading operations. A combustible gas
detector is often used for measurements of percent LEL at potential leak
sources. This instrument is also suitable and recommended for testing
tanks, vessels, and other spaces to determine the presence or absence of
combustible gas or to detect leakage from closed systems.
The required loading rack vapor collection backpressure/vacuum
values can be measured with a pressure gauge. In order to effect this
measurement at the point where the vapor collection hose and the tank
truck interface, a specially designed coupler is employed at the
connection. A tubing connector is situated at the midpoint of the
coupler and tubing is used to allow a reading to be taken with the
pressure gauge. The pressure gauge to be used should have a range of
measurement up to 25 inches water gauge.
3.1.2 Ductwork
Assuming that the ductwork was properly designed to begin with,
the main concerns will be limited to insuring that nothing has been done
to the ductwork that would change the system pressure drop and hence the
system flow. An example would be a rerouting of the original ductwork
design that would result in more bends and/or transition pieces and
hence more system resistance. These changes could also result from
deterioration of the system (i.e., holes, tears, rusting) due to neglect
or inadequate maintenance. The easiest way to inspect the ductwork is
3-6
-------
TABLE 3.1
EFFECTS OF PROBLEMS WITH HOODS
Effect on:
Problem Operation Emissions
Holes in hood. Possible pressure loss in Fugitive emissions in exit
duct. through holes in hood.
Plugged filter in hood. Loss of suction in hood. Emissions escape without
being collected by hood.
Hood poorly positioned to None. Emissions escape without
capture emissions. being collected by hood.
3-7
-------
to follow the individual branches from each hood. Engineering judgment
should be used, as ductwork will often be installed overhead and is not
easily accessible.
Ductwork inspections should pay particular attention to areas
where problems may show up, such as bends, transition pieces, and
dampers. Due to the action of the moving, contaminated air, these
pieces may fatigue more quickly than other components, and must be
thoroughly checked for obvious signs of corrosion and general deteriora-
tion. If a manual damper is installed, the damper should be clearly
labeled with instructions on who is to be notified when it is adjusted.
An out-of-position damper can throw an entire capture system out of
balance, with a resulting decrease in the capture efficiency. A summary
of some problems that may arise with ductwork and the effects that these
problems would have on operations or emissions can be found in Table
3.2.1 A checklist for ductwork that lists the principal items that must
be addressed in a ductwork inspection is presented in Appendix A.
3.1.3 Fans
The fan generates the pressure change in the system that draws air
emissions in through the hoods. If the fan is too small, or operating
too slowly, the airflow will be too low. Most fans encountered in a
plant will be one of two types: direct-driven or belt-driven. Direct-
driven fans are connected directly to the electric motor that drives
them. Fan speeds are limited to available motor speeds, however, and
this makes them very inflexible. Belt-driven fans are connected to
their electric drive motor by means of a belt. This belt gives them the
flexibility of a quick change in fan speed without the need to change
the electric motor. For this reason, belt-driven fans are preferred in
ventilation systems. The volume of air delivered by a fan is directly
proportional to the fan speed; double the fan speed and output volume is
doubled. Belt-driven fans enable plant personnel to adjust the volume
of air they need by adjusting the fan speed.
Unlike hoods and ductwork, fans require regular inspection,
lubrication and maintenance, if they are to perform as designed. Request
the maintenance schedule for the fan, the fan belt, and the motor. Note
whether spare fan belts, fans, and motors are maintained in inventory.
While there may be no regulatory requirements for spare parts, this
information can give clues as to the seriousness of the plant toward
equipment operation and maintenance. Such spares will minimize the
downtime of the capture and control equipment, should an equipment
breakdown occur. A summary of some problems that may arise with fans
and the effects that these problems would have on operations or
emissions can be found in Table 3.3.1 A checklist summarizing items to
be checked during a fan inspection is presented in Appendix A.
3-8
-------
TABLE 3.2
EFFECTS OF PROBLEMS WITH DUCTWORK
Effect on:
Problem
Operation
Emissions
Corroded, eroded metal
ductwork.
Ductwork is vulnerable to
being hit by moving
vehicles.
Buildup of resinous
materials inside duct.
Dampers not properly
labeled (e.g., open,
close).
Dampers open at improper
time.
Pressure loss in duct.
Moving vehicle may damage
ductwork, create holes.
Possible reduction in flow
area or fire if material is
combustible.
Operator may confuse damper
settings.
None.
VOC stream diluted with
ambient air; decreased
capture efficiency
resulting in increased
emissions.
Fugitive emissions from
holes in ducts.
Possible damage to duct
resulting in fugitive
emissions.
Imbalance of airflow in
ductwork reduces capture
efficiency at certain hoods
and results in increased
emissions.
Room air from unused
coating operation dilutes
VOC stream, reducing
capture efficiency and
resulting in increased
emissions.
3-9
-------
TABLE 3.3
EFFECTS OF PROBLEMS WITH FANS
Effect on:
Problem
Operation
Emissions
Fan Motor, bearing
overhearing.
Fan imbalance.
Spare parts (bearing, belt)
not kept in stock.
Belt is slipping, broken.
Damper at fan is not open
wide enough.
Reduced fan efficiency,
suction and VOC capture at
hoods.
Reduced fan efficiency,
suction and VOC capture at
hoods.
System down until spare
part can be obtained.
Fan will not turn at rated
rpm.
Excessive pressure drop
across fan leading to
overheating of motor.
Emissions escape at source
without being collected.
Emissions escape at source
without being collected.
Facility uill most likely
vent emissions to
atmosphere while control
system is down.
Emissions escape at source
without being collected.
Emission escape due to
reduced suction.
3-10
-------
3.2 CONTROL DEVICES
Carbon adsorbers and incinerators, both thermal and catalytic, are
the most frequently installed control devices for the reduction of VOC
emissions. These devices are, therefore, thoroughly discussed in this
section. Condensation and absorption devices can also be used tocontrol
VOC emissions. However, these devices are not likely to be encountered
as frequently in the field. Therefore, operational considerations and
inspection guidance for condensers and absorbers are presented only in
general terms.
3.2.1 Thermal and Catalytic Incineration
Components and Operating Principles.4
The basic purpose of any incinerator is to raise the temperature
of the VOC-containing gas stream sufficiently to allow complete
oxidation of the organic compounds. Thermal incinerators utilize a
burner mounted in the main chamber of the incinerator to generate the
necessary quantity of hot combustion gas. This gas then heats the
relatively cool VOC containing gas stream to a level several hundred
degrees Fahrenheit above the auto-ignition temperature for the specific
organic compound. The auto-ignition temperature is generally in the
range of 800 to 1400 degrees Fahrenheit. In the case of catalytic
incinerators, a preheater burner (or burners) is used to raise the gas
stream temperature to the level necessary to complete oxidation on the
surface of the catalyst bed. Catalytic incinerators generally operate
several hundred degrees below thermal incinerators for the same organic
compounds since the catalyst promotes oxidation reactions. It should be
noted that in both thermal and catalytic incinerators, the VOC compounds
are not oxidized within the burner flame itself. The burner (or
burners) simply provides the turbulent mixing and the hot gas which is
necessary to accomplish VOC oxidation.
Thermal and catalytic incinerator inlet gas stream VOC
concentrations are usually limited to between 500 ppm and 7,500 ppm for
safety reasons. It is generally necessary to maintain VOC
concentrations lower than 25 percent of the Lower Explosive Limit (LEL)
so that the incinerator flame does not flashback to the process
equipment. The 25 percent LEL value is widely accepted upper
concentration limit which allows for some nonuniformity and variability
in the gas stream VOC levels. The concentrations corresponding to 25
percent of the LEL are provided in Table 3.4 for a number of common
organic chemicals. When mixtures of organic compounds are present in
the inlet gas stream, the total concentration is generally limited to 25
percent of the lowest LEL for the various compounds.
Thermal Incinerators. The major components of a thermal
incinerator include a burner, a refractory-lined combustion chamber, and
a stack. The burner includes a combustion air supply controller, a fuel
rate controller, a flashback arrestor, and a burner assembly. A
thermocouple on the discharge of the incinerator is often used to
operate the controller which maintains the proper air/fuel ratio. In
some systems, heat recovery equipment is used on the incinerator
discharge to reduce the operating costs.
3-11
-------
TABLE 3.4
VOC CONCENTRATIONS CORRESPONDING TO 25 PERCENT
OF LOWER EXPLOSIVE LIMITS4
Contaminant Concentration, ppm
Butane 4,750
Ethane 7,500
Ethylene 7,750
Propylene 6,000
Styrene 2,750
Benzene 3,500
Xylene 2,500
Toluene 3,500
Methyl Alcohol 18,250
Isopropyl Alcohol 5,000
Acetone 7,500
Methyl Ethyl Ketone 4,500
Methyl Acetate 7,750
Cellosolve Acetate 4,250
Acrolein 7,000
Cyclohexanone 2,750
Acetaldehyde 10,000
Furfural 5,250
3-12
-------
Most operation and maintenance problems associated with thermal
incinerators concern the burner since this is the component subjected to
the extreme gas velocities and gas temperatures. These problems include
poor fuel atomization (oil-fired units), deposits within the burner
which cause poor air-fuel mixing, inadequate air supply, and quenchingof
the flame on refractory surfaces. Routine maintenance on at least a
quarterly basis is necessary to clean and readjust the burners for
proper operation. Symptoms of poor burner performance include black
smoke generation, lower than normal outlet temperatures, and higher than
normal VOC outlet concentrations.
Thermal incinerators are also subject to problems caused by
rapidly varying VOC concentrations and gas flow rates. These change the
fuel requirements necessary to maintain a stable outlet temperature. A
sudden decrease in the VOC concentrations coupled with an increase in
the gas flow rate can lead to short term periods with lower than
desireable operating temperatures. A sharp increase in the VOC
concentration with a decreased gas flow rate can lead to short term
excursions above the maximum temperature limits of the combustion
chamber.
Catalytic Incinerators. The basic components of a catalytic
incinerator include the preheater burner, a mixing chamber, a catalyst
bed, a heat recovery system, and a stack. The preheater burner is used
whenever supplemental fuel is needed to achieve the necessary operating
temperature. In many cases, the VOC contaminants have sufficient heat
value to achieve the relatively low combustion temperatures without the
preheat burners. Therefore, inspectors should not conclude that the
unit is not operating correctly simply because the preheat burner is not
operating at the time of the inspection. It is quite possible that the
preheat burner is used only during start-up or during periods of low VOC
concentration.
The temperatures required for high efficiency oxidation depend on
the type of catalyst, the incinerator design, and the type of organic
compound.
Catalytic incinerators are vulnerable to a number of operating
problems due to the participation of the catalyst in the oxidation
reactions. These problems include the following:
• catalyst thermal aging;
catalyst burnout due to high temperature fluctuations;
catalyst scouring from catalyst bed;
soot masking of catalyst due to upset combustion conditions
in preheat (oil-fired) burners;
particulate masking of catalyst; and
poisoning of the catalyst by non-VOC contaminants entrained
in the gas stream.
3-13
-------
Thermal aging is the inevitable result of gradual recrystalliza-
tion of the noble metal catalyst materials due to exposure to the hot
combustion products. The catalyst simply becomes less effective in
promoting oxidation of VOC compounds. Because of this problem, all
noble metal catalysts must eventually be replaced with fresh catalysts.
Thermal burnout is the sudden volatilization of the catalytic
compounds from the support matrix which comprises the catalyst bed. The
temperature excursions which cause catalyst losses are often due to an
undesirable increase in the VOC concentration in the waste gas stream.
The catalyst bed must be replaced once significant burnout has occurred.
Masking inhibits catalyst activity by preventing contact between
the vapor phase organic compounds and the surface of the catalyst
material. This can be caused by deposition of particulate material in
the catalyst bed or by soot formation in the preheat burner. Removal of
water soluble materials from the catalyst surface can be accomplished
simply by washing with a detergent solution. Non-water soluble
materials can sometimes be removed by solvent washing and/or physical
scrubbing of the catalyst materials. There is no permanent damage to
the catalyst unless the cleaning process results in physical attrition
of the catalyst from the surface of the substrate.
Poisoning of catalyst involves irreversible chemical reactions
between gas stream contaminants and the catalyst materials. There can
be significant reductions in VOC oxidation efficiency since the affected
catalyst is no longer effective in the oxidation reactions. Therefore,
the catalyst bed must be replaced after a significant fraction of the
catalyst has been affected. It should be noted that the severity of the
impact depends on the specific type of catalyst, the gas stream
temperatures, and the concentration of the catalyst inhibitor.
Compounds which can cause catalyst poisoning include halogens and heavy
metals, such as mercury, antimony, cadmium, etc.
One indication of catalyst inhibition is a lower than "normal" gas
temperature increase across the catalyst bed. Since the oxidation
reactions occurring on the catalyst bed are exothermic, there should be
a significant temperature increase if the catalyst material is in good
condition. Unfortunately, variations in the inlet VOC concentrations can
also affect the gas temperature rise across the bed. Low VOC
concentrations result in a relatively small temperature increase.
Therefore, the inspector must attempt to determine if a small
temperature increase is due to catalyst inhibition or to a short term
decrease in the VOC concentration.
Incinerator System Malfunctions.
Most problems with thermal incinerators concern the burner itself.
These problems include low burner firing rates, poor fuel atomization
(oil-fired units), poor air/fuel ratios, inadequate air supply, and
quenching of the burner flame. Most of these conditions lead to obvious
smoke production and less than desirable gas outlet temperatures. The
adequacy of VOC control is strongly related to the gas outlet
temperature.
3-14
-------
Catalyst deactivation and blinding are the major problems limiting
the performance of catalytic incinerators. Deactivation can be
reversible or irreversible and is caused by the presence of chemicals
which can react with the catalyst materials. Blinding of the catalyst
bed is due to the accumulation of particulate matter on the surfaces of
the catalyst which block the passage of organic vapor to the surfaces.
The gas temperature rise across the bed drops as the catalyst becomes
deactivated or blinded. A summary of effects due to incinerator unit
problems is presented in Table 3.5.1
Inspection of Incineration Devices.4
The inspection procedures for incinerators are commonly classified
in guidance documents as Level 2 and Level 3 inspections. The Level 2
inspection is a detailed walk-through inspection utilizing the on-site
incinerator and process instrumentation. The Level 3 inspection
includes all of the Level 2 steps and also includes the limited use of
portable instruments to verify incinerator performance. The instruments
generally used are the portable VOC detectors and portable thermocouple
thermometers. In some cases, a Level 3 inspection may be warranted if
Level 2 observations indicate a compliance problem. A checklist for
inspection of incineration controls is presented in Appendix A.
Level 2 Inspections
Observe the Incinerator Exhaust. There should be no visible soot
or particulate emissions from the exhaust. Visible emissions are
generally due to improper burner operation or condensation of unburned
organic compounds.
Observe the incinerator bypass stack. Incinerators generally must
have bypass stacks so that the process equipment can be safely vented in
the event of incinerator malfunction. However, during routine
operation, there should be no significant leakage of VOC contaminated
gas through the bypass stack dampers. The leakage of high VOC
concentration gas can often be identified by the wavy light refraction
lines at the stack mouth.
Record the incinerator operating temperature. For thermal
incinerators, the combustion chamber exhaust gas temperature should be
recorded. This is generally monitored by a thermocouple which is used
to adjust the main burner firing rate. A reduction in the operating
temperature could result in a reduced VOC oxidation efficiency.
For catalytic incinerators, the inlet and outlet gas temperatures
across the catalyst bed should be recorded. The inlet gas temperature
is the temperature after the preheater burner and immediately ahead of
the catalyst bed. The bed outlet temperature is the temperature before
the gas stream enters any of the heat recovery equipment. Smaller than
normal temperature increases across the catalyst bed are due to either
catalyst inhibition or to a reduced VOC concentration in the inlet gas
stream.
3-15
-------
TABLE 3.5
EFFECTS OF PROBLEHS WITH INCINERATOR UNITS
Effect on:
Problem
Operation
Emissions
Operating temperature lower
than stated on permit.
Fan malfunction.
Vapor stream not preheated.
Burners plugged, worn.
Incinerator controls not
calibrated.
Incinerator has no
controls.
Improper refractory for
high temperatures.
Fuel filter plugged.
Catalyst plugged, poisoned.
Catalyst not cleaned,
replaced on regular
schedule.
Heat exchanger fouled,
leaking.
Reduced VOC combustion
efficiency.
Loss of VOC; excess airflow
to incinerator.
Additional heat required to
raise vapor stream to
combustion temperature.
Flame out.
Low temperature, high VOC
concentration, fan failure.
Low temperature, high VOC
concentration, fan failure.
Warping, deformation of
metal shell.
Decreased auxiliary fuel
flow.
Inefficient adsorption of
VOC on catalyst surface.
Inefficient adsorption of
VOC on catalyst surface.
Insufficient preheat of VOC
vapor to incinerator.
VOC pass through
incinerator without being
completely oxidized.
Incomplete combustion of
VOC resulting in emissions
higher than design level.
Incomplete combustion.
VOC are not combusted.
VOC not completely
combusted.
VOC not completely
combusted.
Fugitive emissions leaking
from incinerator unit.
Incomplete combustion of
VOC.
Incomplete combustion of
VOC.
Incomplete combustion of
VOC.
Excess emissions due to
incomplete combustion. VOC
may bypass incinerator
through leaks.
3-16
-------
Check for air infiltration into the incinerator system. Air
infiltration into incinerators under negative pressure (fan downstream
of the incinerator) can lead to localized cooling of the gas stream.
Incomplete VOC oxidation can occur in these areas. Severe air infiltra-
tion into the inlet duct could prevent proper incinerator operating
temperatures since this reduces the sensible heat and the heating value
of the inlet gas stream. Infiltration also reduces the VOC capture
effectiveness at the process source. Inspectors can attempt to
audiblycheck for infiltration or employ equipment such as velometers and
smoke tubes to confirm air flow.
Check the incinerator shell, outlet ductwork, and stack for
obvious corrosion. Hydrochloric acid vapor can be formed in inciner-
ators due to the oxidation of chlorinated hydrocarbons. This can lead
to corrosion of the incinerator shell and downstream gas handling
equipment. If this is observed, then the integrity of the system is
suspect and instrumentation should be employed for detection of fugitive
VOCs.
Review the process operating records. Confirm that the inciner-
ator has been operated continuously with the process.
Level 3 Inspections
Measure the VOC outlet concentration. The effluent gas concentra-
tion should be measured if there is safe and convenient access to the
effluent gas duct. The port should be located downstream of the heat
recovery equipment so that the gas temperature is as low as possible. A
glass-lined probe is usually advisable to minimize losses of organic
vapor to the surfaces of the probe. If the gas temperature is greater
than 300*F, it will probably be necessary to include a condenser and
knock-out trap in the sample line in order to protect the teflon tubing
in the VOC detectors.
The VOC detector (and its portable recorder, if any) should be
certified as intrinsically safe for the type of hazardous location
prevailing in the vicinity of the incinerator. No electrically powered
equipment should be used which could conceivably ignite fugitive VOC
vapors.
The observed concentration should be less than 5 to 10 percent of
the inlet concentration if the incinerator is operating properly.
Measure the inlet VOC concentration. The inlet gas stream VOC
concentration can usually be measured using the same VOC instrument used
for the outlet port. A dilution probe will often be necessary for those
photoionization instruments and flame ionization instruments that are
limited to 1000 to 2000 ppm. The condenser and knock-out trap are
rarely necessary since the gas stream temperatures are normally less
than 250*F. As in the case with the outlet measurements, the
measurement of the inlet concentration should be done only when all
safety requirements are satisfied.
3-17
-------
Measure the incinerator outlet temperature. The measurement of
the incinerator outlet temperature is attempted whenever the on-site
gauge does not appear to be providing accurate data. However,
measurement of the outlet temperature using portable gauges is subject
to a number of significant possible errors. These include the
following.
Higher than actual values due to exposure of the probe to
radiant energy from the burner.
Lower than actual values due to shielding of the probe behind
refractory baffles in the combustion chamber.
Non-representative values due to spatial variations of gas
temperature immediately downstream of the incinerator.
For these reasons, the independent measurement of the incinerator
outlet temperature is rarely done by regulatory agency inspectors.
Also, battery powered thermocouple thermometers are not intrinsically
safe and can therefore not be used in certain areas.
Limitations of Present Inspection Techniques.5
The primary consideration in any incinerator system inspection is
an evaluation of the VOC destruction effectiveness. In the case of
thermal incinerators, an adequate evaluation of the incinerator VOC
destruction effectiveness can be made without the aid of portable VOC
analyzers. The outlet gas temperature is a reliable indirect indication
of the overall VOC destruction efficiency (assuming that the integrity
of the ducting and heat exchange system is intact) and most commercial
thermal incinerator systems monitor this temperature continuously.
However, in the case of catalytic incinerators, there is no readily
available operating parameter which an inspector can use to rapidly and
accurately evaluate the adequacy of performance. Inspectors have been
using the gas temperature rise across the catalyst bed in conjunction
with the incinerator outlet temperature. However, it is impossible to
determine whether a low temperature rise is due to catalyst problems,
due to a low VOC inlet concentration, or due to a combination of both
conditions. In some cases, a low gas temperature rise could represent
significant compliance problems, and in other cases it could simply
indicate normal process operating variability. For these reasons, an
alternative inspection technique using portable VOC analyzers would be
helpful in confirming adequate VOC destruction efficiency. Use of
portable VOC analyzers to conduct compliance inspections is discussed in
Section 3.3.
3.2.2 Carbon Bed Adsorbers4
Principles of Operation.
Adsorber units are classified as either non-regenerable or
regenerable. Non-regenerable units are normally used to control exhaust
streams where solvent recovery is not cost-effective. Regenerable units
are used for higher VOC concentrations. Regenerable units are further
subdivided into fixed bed adsorbers, moving bed adsorbers, and fluidized
3-18
-------
bed adsorbers. The predominant regenerative units being fixed beds.
Systems having two or more carbon bed adsorber vessels are typically
used on moderate-to-large process sources such as rotogravure
operations.
A simplified flowchart for a three-bed carbon adsorption system is
shown in Figure 3.2. Process effluent containing VOCs is drawn by a fan
which is usually upstream of the carbon bed adsorber vessels. The
static pressures in the hoods and ductwork leading to the fan are
negative and the static pressures throughout the carbon bed adsorber
vessels are positive when the fan is in the location shown in Figure
3.2. The condition of the hoods and ductwork is obviously important
since any air infiltration upstream of the fan could significantly
reduce the capture of VOC emissions from the process equipment. Any
leakage of VOC contaminated gas after the fan could create unhealthy
organic vapor levels around the equipment.
A particulate filter is used to prevent the deposition of fine
particles on the carbon. By locating it upstream of the fan, the filter
also provides some protection for the fan.
A series of dampers before and after each carbon bed adsorber
vessel are used to direct the VOC contaminated gas stream to the one or
more beds which are in the adsorption mode. The cleaned gas stream is
exhausted directly from the beds to a stack serving the overall system
or to individual stacks with each of the adsorber vessels. Since
adsorption is inherently a batch process, one bed is usually isolated
from the gas stream for desorption of the accumulated organic compounds.
The adsorber vessel operating cycle can be controlled by timers or by
VOC detectors on the outlet of each bed. Generally a desorption cycle
takes an hour and each bed within the system undergoes desorption every
three to six hours of operating time.
During desorption, low pressure steam flows backwards through the
bed. In most cases, the concentrated organic vapors released from the
carbon bed and the steam are processed in a condenser and decanter as
shown in Figure 3.2. However, this is effective only when water-
insoluble organic solvents are present. More complicated solvent
recovery systems are necessary for water soluble organic compounds.
Immediately after desorption, the adsorption vessel is briefly
purged to remove any trapped steam and to cool the bed. This is
necessary to ensure proper adsorption conditions when the adsorption
vessel is brought back on-line.
Carbon Bed System Operating Problems.
When the carbon bed system is operating properly, the emissions
will typically remain in the 50 to 150 ppm range (average emissions)
until the working capacity of one of the adsorption vessels is being
approached. The emissions increase as the bed with the longest on-line
time approaches breakthrough (breakthrough emissions). The emissions
are highest immediately before one bed is isolated for desorption and a
regenerated bed is returned to service. In a properly operating unit,
the overall VOC control efficiency is in the range of 95 to 99 percent.
3-19
-------
Fan
OJ
rvj
o
•Particulate
Filter
Note:
DPI - STATIC PRESSURE DROP
A MOTOR CURRENT
T GAS TEMPERATURE
P PRESSURE
MOC ~ VOC CONCENTRATION
Carbon Bed Adsorber Vessels
Exhaust Gas
Condensor
Steam
Conl
rol Valves
/
i j
Controller
Figure 3.2 Flowchart of a typical multi-bed carbon bed adsorber system.
-------
However, there are a number of operating problems which can result
in substantial decreases in the VOC control efficiency. These problems
include the following:
an increase in the inlet gas temperature and/or the adsorber
vessel operating temperature;
an increase in the feed rate VOC vapor to the adsorber
vessels;
loss of carbon adsorption activity;
deterioration of the physical condition of the carbon bed;
increase in inlet stream moisture content; and
incomplete capture of the VOC emissions from the process
sources.
The first five of the general problems listed above lead to higher
than normal VOC emissions from the carbon beds. Incomplete capture
leads to high fugitive VOC emissions but relatively low emissions from
the carbon bed adsorption vessels. The effects of problems with
adsorption units are summarized in Table 3.6.1
Adsorption of VOCs on carbon is highly sensitive to the
temperature of the carbon. An increase of 15 to 20eF in the inlet gas
temperature can substantially reduce the adsorption capacity of the
carbon for organic compounds. Also, inadequate cooling of the desorbed
carbon bed before returning it to service can lead to reduced carbon
working capacity.
An increase in the feed rate of VOCs to an adsorber vessel reduces
the time period before the working capacity of the bed is reached. This
occurs despite the fact that the working capacity is increased slightly
as the VOC concentration increases.
Loss of carbon bed activity occurs primarily due to the gradual
accumulation of high molecular weight organic compounds which can not be
effectively removed during the relatively mild desorption conditions.
These high molecular weight compounds are often introduced to the system
as contaminants in the solvent. Incomplete desorption can also lead to
reduced bed activity.
Physical deterioration of the carbon bed can occur due to the
collapse of the retention screens holding the carbon pellets. Corrosion
of these screens is common. The deposition of particulate matter can
also damage the beds.
Carbon bed operating problems result in VOC concentrations
substantially above the 50 to 150 ppm range. The maximum effluent
concentration is limited only by the inlet gas stream VOC concentration
which is normally 2,000 to 7,500 ppm.
3-21
-------
TABLE 3.6
EFFECTS OF PROBLEMS WITH ADSORPTION UNITS
Problem
Effect on:
Operation
Emissions
Regeneration concentration
detect ion/initiator poorly
calibrated.
Gradual loss of adsorbent
due to entrainment.
VOC inlet vapor
concentration higher than
design value.
Emergency bypass damper
opened.
Regeneration steam traps
not bled.
Adsorbent not inspected,
replaced on regular
schedule.
Internal vessel liner
chipped, abraded.
Plugged prefilter.
Blower (fan) failure (e.g.,
bearings, belt, motor).
Inlet VOC stream relative
humidity >50 percent.
Nonregenerable compounds
present in inlet VOC
stream.
Corrosion in adsorber.
Vapor stream inlet
temperature higher than
design value.
Inlet vapor stream relative
humidity <20 percent.
Regeneration will be
initiated improperly.
Loss of adsorption capa-
city, premature saturation
of bed.
Adsorbent is prematurely
saturated. Also, possible
hot spots and bed fires may
occur.
Loss of stream to adsorber.
Steam becomes saturated
(wet).
Buildup of particulates,
nonregenerable organics on
adsorbent may occur.
Eventual corrosion, erosion
of vessel walls.
Excessive pressure drop in
vapor line.
Reduced flow to adsorber.
Water vapor competes with
VOC for adsorption sites.
Bed fouling, less adsorp-
tion capacity.
Pressure loss in bed.
Revaporization of low
boiling compounds will
occur.
Potential excess heat
buildup in bed.
Possible premature
breakthrough.
Breakthrough (emission of
uncaptured VOC) occurs much
sooner.
Premature breakthrough.
Emissions bypass control
unit and are vented to
atmosphere.
Bed is saturated with water
and cannot adsorb VOC.
Premature breakthrough.
Fugitive emissions exit
through vessel walls.
Increase in bed exhaust
concentrations.
Increase in bed exhaust
concentrations.
Breakthrough occurs much
sooner.
Breakthrough occurs much
sooner.
Fugitive emissions from
adsorber vessel.
Exhaust VOC concentrations
higher than design value.
Breakthrough occurs much
sooner.
3-22
-------
TABLE 3.6 (Continued)
EFFECTS OF PROBLEMS WITH ADSORPTION UNITS
Problem
Effect on:
Operation
Emissions
Highly exothermic solvents
(e.g., ketones, phenols)
present in inlet VOC
stream.
Condensibles transfer pump
leaking.
Control system poorly
maintained.
Hot spots, bed fires may
occur resulting in adsorp-
tion capacity loss.
Leaking condensed VOCs.
Poorly controlled system.
Exhaust VOC concentrations
in excess of emission
limits.
Vaporization of leaked
VOCs.
Potential excess concentra-
tion of VOC in adsorber
exhaust stream.
3-23
-------
Inspection of Carbon Bed Adsorbers.4
Techniques for the inspection of carbon bed adsorbers are commonly
classified in guidance documents as Level 2 or Level 3. The Level 2
inspections primarily involve a walk-through evaluation of the carbon
bed adsorber system and process equipment using on-site gauges. The
Level 3 inspection incorporates all of the inspection points of the
Level 2 inspection and includes independent measurements of the adsorber
operating conditions. In some cases, a Level 3 inspection may be
warranted if Level 2 observations indicate a compliance problem. A
checklist for inspection of carbon bed adsorbers is presented in
Appendix A.
Level 2 Inspections
Evaluate the VOC outlet detector. The VOC detectors often used at
the outlet of the carbon bed systems are relatively sophisticated
instruments which require frequent maintenance. Confirm that they are
working properly by reviewing the calibration records since the previous
inspection. Maintenance work orders should also be briefly reviewed to
determine if the instruments have been operational most of the time.
Check carbon bed shell for obvious corrosion. Some organic
compounds collected in carbon bed systems can react during steam
regeneration. This leads to severe corrosion of the screens retaining
the carbon beds and of the unit shell, and could signal problems
requiring that a Level 3 inspection be conducted.
Observe the adsorption/desorption cycles. Determine the time
interval between bed regenerations and compare this with previously
observed values. An increase in this time interval could mean that
breakthrough is occurring if the quantities of VOCs entering the carbon
bed have remained unchanged. Systems in which the cycle frequency is
controlled by a timer rather than an outlet VOC detector are especially
prone to emission problems due to longer than desirable cycle times.
Evaluate carbon bed system static pressure drop. If there are on-
site gauges, evaluate any changes in the static pressure drop. A
decrease could mean deterioration of the carbon bed to the point that
channeling of the gas stream is affecting gas-solid contact. Higher
than normal static pressure could mean partial pluggage of the carbon
bed due to fines formation or due to material entering with the gas
stream. However, gas flow changes could also be responsible for changes
in the static pressure drop.
Prepare solvent material balances. For some processes, the
effectiveness of the carbon bed system can be evaluated by preparing a
solvent material balance around the facility for a period of several
weeks to a month. The information needed for the calculations includes
solvent quantities purchased, changes in solvent storage tank levels,
and solvent quantities transferred from the system.
3-24
-------
Level 3 Inspections
Measure the VOC outlet concentrations. The effluent concentration
from each bed should be measured if there is safe and convenient access
to the effluent ductwork. The instrument (and its portable recorder, if
any) should be certified as intrinsically safe for Class I, Group C and
D locations. This means simply that the instrument is incapable of
initiating an explosion when used properly. A small port is adequate to
draw a 0.5 to 3.0 liter per minute sample into the instrument.
It is important to determine the approximate desorption cycle of
multi-bed systems. Outlet VOC measurements conducted earlier in the
adsorption cycle of a bed may appear adequate even when the bed activity
is severely reduced. Breakthrough usually does not occur until late in
the operating cycle unless the condition of the carbon adsorbent is
extremely poor. Therefore, an effort should be made to measure the
outlet VOC concentration of each bed at a time when it is approaching
the end of the adsorption mode. The adsorption/desorption cycle is
normally controlled by a timer and this can be used to determine the
approximate status of each bed.
In some commercial multi-bed units there is at best poor
accessibility to the effluent ducts from each unit. In this case, the
VOC concentration in the combined duct should be measured at the exhaust
point. Obviously, this measurement should be attempted only when there
is safe and convenient access to the exhaust. It is especially
important to avoid areas where high VOC concentrations could accumulate.
Measure the inlet gas temperature. Adsorption is inversely
related to the gas temperature entering the carbon bed adsorbers. An
increase in the gas temperature from the baseline period could result in
a decreased capacity for VOCs. The gas temperature should be measured
in the inlet ductwork, immediately ahead of the carbon bed.
Measure the static pressure drop. A change in the static pressure
drop since the baseline period is usually due to either a change in the
gas flow rate through the carbon bed or due to the physical
deterioration of the bed itself. Measurement taps on the adsorber shell
should be used, if available. Alternatively, the static pressure drop
can be measured using ports in the inlet ductwork to the adsorber system
and the outlet duct from the adsorber. Obviously, the static pressure
drop should be determined while the adsorber is on-line.
Limitations of Present Inspection Techniques.4
The inspection of carbon bed adsorbers has previously been limited
to an evaluation of adsorber system operating parameters or a
compilation of a complete material balance. Neither approach has been
entirely satisfactory.
Preparation of the solvent balances involves a review of numerous
plant records for a period of a week to one month concerning solvent
purchases, solvent disposal, and solvent usage. Measurement of the
solvent inventories in storage must be done immediately before and after
the period of interest. Despite the substantial work required in
3-25
-------
preparation of the solvent material balance, the results are not a
sensitive indicator of solvent collection efficiency. Furthermore, it
is impossible to determine if the estimated solvent losses are due to
poor carbon bed adsorber performance or due to inadequate capture of the
solvent vapors at the process equipment.
Evaluation of carbon bed system operating parameters can be done
during a routine Level 2 inspection. However, most carbon bed systems
have only limited instrumentation. Typically continuous measurements
are taken of the following:
inlet gas stream temperature;
steam pressure;
carbon bed adsorber outlet temperature;
condenser water temperature; and
carbon bed inlet and outlet static pressure.
The data provided by these instruments does not provide a clear
indication of VOC emissions from the carbon bed system. It is possible
to have high VOC emissions even when there are relatively normal
pressure and temperature readings. These instruments are provided by
the manufacturer simply to help the plant operations personnel to
operate and maintain the units.
Some systems have a VOC analyzer on the stack and/or the outlet of
each carbon bed adsorber vessel. When these are working properly, they
provide very relevant data for the inspector. However, it is difficult
to maintain these sophisticated instruments in the moderately hostile
environment existing in the carbon bed outlet ducts. For this reason,
inspectors should not use an evaluation technique that depends primarily
on these on-site instruments. An alternative inspection technique using
portable VOC analyzers would be helpful in confirming adequate VOC
control efficiency. The use of portable VOC analyzers to conduct
compliance inspections is discussed in Section 3.3.
3.2.3 Absorbers (Scrubbers)
Absorbers are not likely to be encountered as frequently in the
field as incinerators and adsorbers; therefore, the following
operational considerations and inspection guidance are presented only in
general terms.
Absorption refers to the transfer of a gaseous component from the
gas phase to a liquid phase. With regards to VOC control, absorption
involves the removal of gaseous contaminants from a process exhaust
stream by dissolving them in a liquid stream. Absorption devices are
generally referred to as "absorbers" or "wet scrubbers." The liquid
into which the contaminant is absorbed is referred to as the
"absorbent," while the gaseous contaminant being absorbed is referred to
as the "solute." Common adsorbents for organic vapors are water, non-
volatile organics, and aqueous solutions.
Absorbers are sometimes used in combination with adsorption
devices. An example would be the carbon adsorption - absorption type
vapor recovery systems employed at bulk gasoline terminals. In this
3-26
-------
case, gasoline vapor is the solute and liquid gasoline from storage acts
as the absorbent. This type of control configuration results in product
recovery, and is most effectively employed at process operations with a
uniform vapor composition.
Parameters which are typically monitored on absorbers are:
absorbent flow rate;
absorbent pressure;
pressure drop across control device;
inlet gas temperature; and
outlet gas VOC concentration.
Design considerations for absorbers used to control gaseous
emissions are:6
providing a large interfacial contact area;
providing good mixing between gas and liquid phases;
allowing sufficient residence or contact time between the gas
and liquid phase; and
ensuring a high degree of solubility of the contaminant in
the absorbent.
Additionally, in selecting the absorbent, the following
requirements need to be considered:
The solubility of the gas in the absorbent should be
relatively high so as to enhance the rate of absorption and
decrease the quantity of absorbent required. An absorbent
chemically similar to the solute generally provides good
solubility;
Relatively low in volatility to reduce absorbent cost;
The absorbent should be non-corrosive, if possible, to reduce
construction material cost of the absorber;
• Inexpensive and readily available;
Relatively non-toxic, non-flammable, chemically inert, and
low freezing point.
Factors affecting absorption efficiency are temperature,
concentration of gas, and liquid-to-gas ratios. The lower the
temperature, the higher the absorption. Absorption is increased by
higher gas concentration, as with higher liquid-to-gas ratios. The
effects of some common problems with absorption units are summarized in
Table 3.7.1
3-27
-------
TABLE 3.7
EFFECTS OF PROBLEMS WITH ABSORPTION UNITS
Effect on:
Problem
Operations
Emissions
Incorrect solvent used in
absorber (e.g., water used
to absorb VOC).
Plugged prefilter on vapor
inlet.
Fan failure.
Plugged solvent
recirculation line
strainer.
Solvent recirculation pump
leaking or overheating.
Bypass damper open.
Solvent spray nozzles
plugged.
Packing trap plugged or
fouled.
Stripping column
temperature too low.
Solvent recirculation line
not bled frequently enough.
Incomplete absorption of
immiscible vapor.
Excess pressure drop in
vapor inlet line.
Reduced vapor flow through
vessel, poor gas/liquid
contact.
Reduced liquid flow to
absorber.
Reduced liquid flow to
absorber.
Vapor stream bypasses
absorber.
Poor solvent distribution
in absorber.
Excess pressure drop
through vessel, poor
gas/liquid contact.
Absorbed VOC not fully
desorbed from solvent.
Buildup of VOC in solvent
results in reduced solvent
absorption capacity.
VOC pass through bed with
negligible control.
Possible increase in
exhaust VOC concentrations.
Exhaust VOC concentration
exceeds design limit.
Exhaust VOC concentration
exceeds design limit.
Exhaust VOC concentration
exceeds design limit.
VOC vented directly to
atmosphere.
Exhaust VOC concentration
exceeds design limit.
Exhaust VOC concentration
exceeds design limit.
Exhaust VOC concentration
exceeds design limit.
Exhaust VOC concentration
exceeds design limit.
3-28
-------
3.2.4 Condensers6
Condensers are not likely to be encountered as frequently in the
field as incinerators and adsorbers; therefore, the followingoperational
considerations and inspection guidance are presented only in general
terms.
Condensation is the process of reducing a gas or vapor to a
liquid. Even though condensation can be performed by an increase in
pressure, the most common approach for reducing a gas/vapor to a liquid
is by reducing the temperature. Normally the gas/vapor stream
temperature is reduced by water or air. Other coolants include chilled
brine and refrigerants.
Condensers fall into two basic categories - contact and surface
condensers. In a contact condenser the coolant and the vapor stream are
physically mixed. They leave the condenser as a single exhaust stream.
In surface condensers the coolants and the vapors are separated by a
non-permeable medium which transfers the temperature difference between
the coolant and the vapor. Surface condensers are further subdivided
into single-pass and multiple-pass units. In single pass units the
coolant and vapor only pass each other once. Multiple-pass units employ
two or more vapor and coolant passes to achieve condensation.
Condensers are generally used on streams having a high
concentration of VOC such as in the case of the desorption cycle of a
carbon adsorber. Condensers are not typically used when VOC
concentrations fall below 25 percent of the Low Explosive Limit (LEL).
Parameters which are typically monitored on condensers include the
inlet and outlet temperatures of the coolant and vapor streams.
Sometimes the VOC content of the gas outlet vent is monitored,
especially if it is vented directly to the atmosphere. When condensers
vent directly to the atmosphere, refrigeration of the coolant is
necessary to obtain the vapor removal required by most regulations.
3.3 INSPECTIONS WITH PORTABLE VOC ANALYZERS
The EPA Stationary Source Compliance Division (SSCD) has sponsored
several studies regarding the feasibility of employing portable VOC
analyzers to enhance inspections of VOC control systems at stationary
sources. Considering that these portable VOC analyzers were designed
primarily for fugitive VOC assessment and not for VOC control system
inspection, a number of measurement errors and instrument operating
problems are possible.
Generally, regulatory field personnel will not employ portable VOC
analyzers in an inspection of incineration devices; however it should be
noted that there are a number of potential problems which can limit the
use of portable VOC analyzers in this type of application.5 These can
include:
Inadequately cooled stack gas can damage the portable VOC
instruments' detectors and/or sample lines.
3-29
-------
Condensation of nonvolatile organic compounds can plug the
instruments' particulate filters or damage the instruments'
detectors.
The necessary sampling train may be bulky to transport and
time consuming to set-up.
A time consuming traverse of the stack may be necessary to
avoid measurement errors due to VOC vapor stratification.
• The "specific organic compounds present in the stack are not
known. Therefore, the response factors can not be determined.
The first three problems listed could conceivably be minimized by
the redesign of the sampling probe and by the use of modified sampling
procedures. The last two problems may or may not be significant
depending on site specific incinerator operating conditions. While
inspections done on catalytic incinerators using portable VOC analyzers
could be superior to those done by means of incinerator operating
parameter evaluation, inspectors must recognize that these instruments
do not provide a direct indication of outlet VOC concentrations. This
limitation is due to the unknown response factors. Portable VOC
analyzers could be used to measure the VOC concentrations at a location
upstream of the incinerator. A significant decrease in the VOC
concentration at equivalent production rates and typical solvent usage
rates provides direct evidence of capture problems. Furthermore, VOC
concentration measurements made at these locations are relatively simple
due to the low gas temperatures which often exist upstream of the
incinerators.
Portable organic vapor analyzers could conceivably be used to
enhance the inspections of carbon bed adsorbers. These instruments are
especially attractive since the stack (or adsorber exit) VOC
"concentration" is a direct indicator of the carbon bed system
performance. Furthermore, the inlet VOC concentration of the carbon bed
system can be measured if there is any question regarding the adequacy
of hood capture or regarding the extent of air infiltration into the
ventilation system. Low values may indicate fugitive emissions
problems. Even the maximum design VOC concentrations are within the
measurement range of most portable VOC analyzers when dilution probes
are used. However, small measurement ports would have to be installed
on some commercial systems.
Potential moisture problems can be encountered in applying
portable VOC analyzers to carbon adsorption systems. After the
desorption cycle, the adsorber vessel is purged of the steam trapped
inside. In some systems, this gas does not pass through the condenser.
Therefore, steam emissions exist for a short period of time. It is
especially important that the portable instruments not be used during
* These instruments do not provide a direct measure of the actual VOC
concentration unless they have been calibrated with the specific compound
or mixture of compounds present in that particular emission stream.
3-30
-------
this period since even small quantities of moisture can temporarily
impair all of the major types of portable VOC analyzers.
Since carbon bed systems generally handle only one or two types of
solvents, the portable VOC analyzers can be set for these specific
compounds. This reduces the instrument data interpretation problems
discussed earlier.
3-31
-------
3.4 REFERENCES
1. Guide for Inspecting Capture Systems and Control Devices at
Surface Coating Operations. Draft Final Report, GCA
Corporation. EPA Contract No. 68-01-6316, Task Order No. 33.
May 1982.
2. McDermott, H.J. Handbook of Ventilation for Contaminant
Control. Ann Arbor Science. Ann Arbor, Michigan. 1979.
3. ACGIH Committee on Industrial Ventilation. Industrial
Ventilation. A manual of recommended practice, 14th Edition
Lansing, Michigan: American Convergence of Governmental
Industrial Hygienists. 1976.
4. Field Inspection Notebook. Richards Engineering, Durham, NC.
Purchase Order 6D3843NASA. Draft Version. Prepared for U.S.
Environmental Protection Agency, Air Pollution Training
Institute. August 28, 1987.
5. Development of VOC Inspection Instrumentation Guide. U.S.
Environmental Protection Agency, Washington, DC. Contract No.
68-02-3962, Work Assignment 3-89. Richards Engineering, Durham
NC. January 1987.
6. EPA Region VI Air Compliance Determination Vol. I: Overview.
Glenn G. Draper Engineering, Dallas, TX. Contract No. 68-02-
4465, Work Assignment No. 008. Prepared for U.S. Environmental
Protection Agency, Region VI, Air Enforcement Division.
September 1987.
3-32
-------
CHAPTER 4
CONCLUSIONS AND RECOMMENDATIONS
A continuous monitoring system, properly operated, can provide
source personnel and inspectors alike with pertinent continuous
operating data concerning the operation of capture and control systems.
In selected applications, a monitoring system can go beyond performance
indications and serve as a continuous compliance monitor. Alterna-
tively, due to variances in emission stream characteristics and
instrument limitations, not all sources lend themselves to continuous
compliance monitoring.
The discussion alternatively uses terms such as continuous
monitoring, continuous emission monitoring (CEM), and continuous
compliance monitoring (CCM). As a point of clarification, the term CEM
is used to denote a VOC detection system, such as an NDIR, employed as a
performance indicator. The term CCM is used to define a monitoring
system used as an indicator of regulatory compliance.
4.1 GUIDELINES FOR APPLICATION OF CONTINUOUS MONITORS
In the following paragraphs, applications of the monitoring
systems identified are discussed. Specific process and control system
parameters to which these monitors may be applied, are categorized per
the areas presented in Section 2.1 (e.g., monitoring of incineration,
solvent recovery and capture systems). Also presented are some of the
potential problems and limitations identified.
4.1.1 Incineration Operations
Monitoring requirements identified for thermal and catalytic
incineration (Section 2.1.1) pertain only to continuous temperature
measurement devices. This requirement is based upon the premise that
combustion chamber temperature is the fundamental indicator of VOC
destruction efficiency. These devices, where required, are to be
equipped with recording devices so that permanent, continuous records
may be produced.
Temperature Monitoring
Many State and local agencies routinely issue permits requiring
that monitoring of incineration temperature be conducted. The most
frequently mentioned aspect to ensure satisfactory performance of
4-1
-------
temperature monitoring systems involves correct placement of the
thermocouples including use of shields to prevent effluent "backwash",
which may cause readings lower than the actual value. Additional
considerations include use of the correct thermocouple type, and
frequent calibration of the devices. In actual practice, a thermocouple
is often used on the discharge of the incinerator to operate the
controller which maintains the proper air/fuel ratio. However, these
devices are not typically equipped with continuous recording devices.
In this sense, the temperature monitoring is conducted more for
performance enhancement than for compliance purposes. While these
existing monitoring systems could be equipped with continuous recording
devices, the obvious drawback is the cost to install, calibrate,
operate, and maintain such a system. Another drawback to the use of
continuous temperature monitoring for compliance is that, although
certain relative accuracies are required, no procedures are presented
for their determination. Therefore, when selecting a temperature
monitoring device, a temperature range should be chosen where accuracy
is deemed the most critical. This is due to the fact that the
thermocouple exhibits somewhat poor accuracy when measuring temperatures
over a wide range.
VOC Concentration
VOC CEMs are not typically required as monitoring systems for
incineration units. In fact, no specific application of VOC CEMs to
incineration units was identified. If required, however, no operational
constraints for their application were identified. Monitoring should be
conducted downstream of heat recovery equipment so that the gas
temperature is as low as possible. If the gas temperature is greater
than 300*F, the sample line will probably need to include a condenser
and knock-out trap in order to protect the teflon tubing in the VOC
detectors. The equipment should be certified as intrinsically safe, due
to potential ignition of fugitive VOC vapors.
Under normal catalytic incinerator operating conditions, there is
a significant temperature increase across the catalyst bed. However,
variations in the inlet VOC concentrations can also affect the
temperature rise across the bed. Low VOC concentrations result in a
relatively small temperature increase. Therefore, a determination must
be made as to whether a small temperature increase is due to catalyst
inhibition, due to a short-term decrease in the VOC concentration, or
due to a combination of both conditions. For these reasons, an
alternative inspection technique using continuous or portable VOC
analyzers would be helpful in confirming adequate VOC destruction
efficiency. The inlet gas stream VOC concentration can usually be
measured using the same VOC instrument used for the outlet port. A
dilution probe will often be necessary for photoionization instruments
and flame ionization instruments that are limited to 1000 to 2000 ppm.
The condenser and knock-out trap are rarely necessary since the gas
stream temperatures are normally less than 250°F.
Pressure Monitoring
For those VOCs introduced into the combustion chamber, monitoring
of combustion temperature provides a good indication of incinerator
4-2
-------
performance. However, additional parameters may need monitoring to
ensure that all VOCs within the emission stream do, in fact, reach the
chamber. Loss of pressure across a heat exchanger or catalyst bed may
indicate a decrease in VOC emission reduction.
The most common use of a heat exchanger in a VOC incinerator
installation is for preheating of the VOC feed stream. A breakdown of
the heat exchanger can cause the release of unburned VOC by allowing the
VOC feed stream to enter the incinerator exhaust, thereby bypassing the
incinerator.
Pressure drop across the catalyst bed of an incinerator unit also
provides an indication of performance. A high pressure drop would
indicate blinding or clogging of the catalyst, while a low pressure drop
might be indicative of gas stream channeling.
Common static pressure gauges may be used to evaluate the pressure
drop across a heat exchanger or a catalyst bed. For a discussion of
pressure monitoring systems, refer to page 2-10.
4.1.2 Solvent Recovery Operations
Requirements identified for processes controlled by solvent
recovery devices (e.g., carbon adsorbers and condensers) can be grouped
into the following three categories of compliance monitoring: 1)
measurements of solvent recovered, 2) VOC concentration level
monitoring, and 3) monitoring of exhaust gas stream temperature.
Solvent Recovery Monitoring
NSPS requirements state that a solvent recovery monitoring device
must be accurate within ± 2.0 percent. A survey of regulatory personnel
and equipment vendors revealed that no such type of continuous
monitoring device was being used for compliance purposes. Gauging
equipment, consisting of meters or strain gauges, is commonly used today
in many industrial settings to measure cumulative amounts of liquids
contained in tankage. Hard copies of continuous readouts are not
routinely obtained; however, operations personnel may make note of
liquid level^ at regular time intervals.
VOC Concentration
Since adsorption is inherently a batch process, one bed is usually
isolated from the gas stream for desorption of the accumulated organic
compounds. The adsorber vessel operating cycle can be controlled by
timers or by VOC detectors on the outlet of each bed. In fact, many
systems have a continuous organic vapor analyzer at the stack for this
purpose. However, while these analyzers provide a good indication of
performance; in many cases, they cannot be used as an indicator of
compliance. This is mainly due to process variability-related problems
and instrument limitations. Due to the operating principles of VOC
CEMs, VOC compounds in the emission stream must remain relatively
constant over time to ensure the instrument is calibrated to current
stream conditions. Numerous source types possess emission stream
4-3
-------
characteristics that change radically over the course of a day. These
problems are discussed in detail within Section 2.3.
When applying VOC CEMs to specific sources, it is important to
remember that exposure of these systems to certain environments
significantly degrades performance. Sample conditioning systems are
very important in applications where temperature, particulates, and
moisture could present a problem. One such problem was identified where
application of a VOC CEM and sample conditioning system to a rotogravure
printing operation was improperly engineered. A constituent of the
printing inks would "gum up" the sample line resulting in chronic
plugging problems.
Temperature Monitoring
Adsorption of organic vapor on carbon is highly sensitive to the
temperature of the carbon. An increase of 15 to 20'F in the inlet gas
temperature can substantially reduce the adsorption capacity of the
carbon for organic compounds. Also, inadequate cooling of the desorbed
carbon bed before returning it to service can lead to reduced carbon
working capacity. Continuous monitoring of these parameters, although
not useful as a compliance measure, would provide a useful indication of
performance when compared to established baseline values.
In cases where condensers are used for control, monitoring of gas
stream temperature would provide support for compliance. Parameters
which are typically monitored on condensers are the inlet and outlet
temperature of both the coolant and the vapor stream. As previously
stated, such a monitoring system could be equipped with a continuous
recorder. However, if the condenser effluent is exhausted directly to
the atmosphere, then monitoring of gas stream VOC content may also prove
necessary for compliance determination.
Pressure Monitoring
The carbon bed system static pressure drop provides an indication
of performance. However, while changes in the static pressure drop
could be due to physical deterioration of the carbon bed, these
fluctuations could also be caused by variations n gas flow to the bed.
4.1.3 Operation of the Capture System
An affected facility which achieves compliance with the use of
add-on control equipment, including a VOC capture system, is required by
certain NSPS regulations (Subparts RR and SSS) to submit a monitoring
plan for the VOC capture system. This plan is to identify the parameter
to be monitored as an indicator of VOC capture performance (e.g., the
power supply to the exhaust fans or duct flow rates) and the method for
monitoring the chosen parameter. The facility is required to install,
calibrate, maintain, and operate, a monitoring device that continuously
indicates and records the value of the chosen parameter.
None of the EPA, State or local agency personnel contacted could
cite a case where continuous monitoring has been applied to any capture
system operating parameters. Typically, these parameters are visually
4-4
-------
monitored during compliance testing of the VOC control system in order
to establish a baseline and comparative basis for future observations.
One printer was identified that employs on-site LEL monitors to
ensure that explosive conditions are not present. The monitors were
continuous FID units, set to alarm at 40 percent LEL and effect unit
shutdown at 50 percent LEL. In addition to safety, the LEL monitors
also enhance process economics by adjustment of air flow recirculation
in order to reduce flow to the incinerator. Typically, insurance
requirements and the facilities themselves dictate installation of an
LEL monitoring system, and air regulations are not required to address
their implementation.
Portable VOC analyzers were discussed as an inspection tool.
While these instruments do not provide a direct indication of VOC
"concentration," they can provide direct evidence of capture problems.
A detailed discussion was provided in Section 3.3.
4.2 CONTINUOUS COMPLIANCE MONITOR PERFORMANCE GUIDELINES
As discussed in Section 2.2, three types of VOC CEM/CCM systems
are employed in industrial processes: FIDs, PIDs, and NDIRs.
Presently, no guidelines or specifications for performance of these
systems have been established. Consequently, some State and local
agencies are developing, on a case-by-case basis, performance
specifications for VOC CCMs upon those established for CCM systems which
monitor other pollutants. Specifically, the guidelines presented in
Performance Specification 2 (PS 2K as detailed in 40 CFR 60, Appendix B
are used as a basis for VOC CCMs. ' The guidelines presented in PS 2
were written specifically for S02 and N0x CCM systems employed on
stationary sources. Nonetheless, the specifications provided for these
systems are applicable to VOC CCM systems. The guidelines presented in
PS 2 concentrate upon system calibration, in particular upon relative
accuracy and calibration drift.
Table 2.3 presented detailed performance specifications for an
existing VOC CCM system monitoring gasoline bulk loading operations.
These specifications do not identify specific calibration techniques nor
are they designed to evaluate CCM performance over an extended period of
time. It is often up to the regulatory agency to determine calibration
techniques and ensure the CCM systems perform satisfactorily over time.
In addition, NSPS Subpart RR requires EPA, State and local agencies to
define a methodology to ensure VOC CCM relative accuracy to ± 2 percent
when determining the amount of solvent recovered. It does not appear
that is possible to define such a methodology for this purpose on a
broad scale given the complexity of some sources. It would be more
feasible to define such a methodology for sources exhibiting uniform
emission characteristics. A further discussion on this is provided in
Section 2.3.1.
If a VOC CCM is to be required as part of a rule, then it may be
necessary to revise the existing VOC emission limitation to allow the
instrument direct measurement of the compliance parameter. For example,
in instances where the regulations limit emissions on a Ib/hr basis,
4-5
-------
revising the regulation to a concentration limitation will enable the
VOC CCM to directly measure the compliance parameter. Refer to section
2.3.1 for further discussion.
It is important to remember that correct instrument calibration
and relative accuracy are only a part of ensuring that monitor
performance is satisfactory. Maintenance on the system must be done to
keep performance standards high. In general, maintenance requirements
on VOC CEM/CCMs set forth by regulatory authorities recommend following
manufacturer's guidelines. Sources can choose monitoring systems from a
variety of manufacturers, and often each manufacturer recommends
different maintenance procedures. It is therefore difficult for
regulatory agencies to specify maintenance requirements beyond those
outlined by the manufacturer. In light of this, regulatory agencies
should request a copy of the maintenance procedures recommended by the
manufacturer and ensure these steps are followed.
To assist EPA, State and local agencies in determining VOC CEM/CCM
system performance, portable VOC analyzers may prove very useful during
periodic site visits or inspections. Due to variations between VOC
analyzers, the reading obtained from a portable unit will likely not
match the system unit reading precisely. Nonetheless, the readings
should be within an order of magnitude of one another. Refer to Section
3.3 for a more complete discussion on portable VOC analyzers.
4.3 RECOMMENDATIONS AND CONCLUSIONS
The following items detail recommendations arising from this
study.
EPA should develop performance specifications for VOC CCMs
based on those detailed in Performance Specification 2, 40 CFR
60, Appendix B; currently no performance specifications for
VOC CEMs/CCMs exist. In addition, the NSPS regulations may
need to be revised in order to implement VOC CEMs as a
compliance tool. VOC emission limitations given in NSPS
regulations typically are stated in mass emission rates of VOC
(e.g. Ib VOC/hr). Other VOC emission limitations given in
NSPS regulations specify a control efficiency for add-on VOC
control devices. VOC CEM/CCM instruments directly measure
mass VOC concentration not emission rates. Thus, NSPS
regulations limiting mass emission rates may require revision
if VOC CCMs are to be implemented.
EPA should develop a methodology to ensure relative accuracy
to ± 2 percent for solvent recovery monitoring instruments
specified in NSPS Subpart RR.
Some States and local agencies have developed VOC CEM
specifications based upon Performance Specification 2 (PS 2),
as detailed in 40 CFR 60, Appendix B.
Flame ionization detectors (FIDs), photoionization detectors
(PIDs), and non-dispersive infrared detectors (NDIRs) may be
4-6
-------
used as a CCM in some instances, but instrument limitations
and process complexities limit their use on a broad scale.
Sources operating VOC CCMs should forward a copy of vendor
maintenance requirements to regulatory agencies, who, in turn,
should ensure that the sources follow these requirements.
Sources having more consistent gas stream characteristics are
better candidates for use of VOC CEMs as compliance monitors
(VOC CCMs).
Where VOC CEM/CCMs are applied, careful consideration should
be given to engineering of the sample conditioning systems.
Use of a portable VOC analyzer by regulatory agencies and sources
is encouraged to assist in determinations of VOC CEM/CCM system
performance, as well as to determine the integrity of the VOC
capture system.
Continuous temperature monitors provide an accurate indication
of compliance for thermal incinerators.
In addition to continuous temperature monitoring, an
alternative inspection technique using continuous or portable
VOC analyzers would be helpful in confirming adequate VOC
destruction efficiency in catalytic incinerators.
Incineration systems (both thermal and catalytic), equipped with
heat exchangers should be required to continuously monitor pressure
drop across the exchanger to ensure that the combustion chamber is
not bypassed.
Process lines using air flow recirculation are usually required to
employ an LEL indicator. Their installation is usually driven by
insurance requirements and safety concerns.
4-7
-------
4.4 REFERENCES
1. Telecon. Sink, M., Pacific Environmental Services, Inc. with
Jim Nolan, Puget Sound Air Agency. June 20, 1989.
2. Telecon. Sink, M., Pacific Environmental Services, Inc. with
Bill Chaffin, Texas Air Pollution Control Board. July 10,
1989.
4-8
-------
APPENDIX A
COMPLIANCE INSPECTION CHECKLISTS
A-l
-------
TABLE A.I
CHECKLIST FOR HOODS
PLANT XYZ. INSPECTOR -J •
DATE
I. PROCESS
A. PROCESS LINE DESIGNATION
Q /,
B. PROCESS EQUIPMENT CONTROLLED BY HOOD
C. PROCESS IN OPERATION DURING INSPECTION? YES V/ NO _
D. 1. PROCESS OPERATING AT MAXIMUM CAPACITY? YES V NO _
2. IF NOT, AT APPROXIMATELY WHAT PERCENT OF CAPACITY? _ X
II. HOOD
A. TYPE OF HOOD: ENCLOSURE RECEIVING
CAPTURE V/ OTHER (DESCRIBE)
B. 1. IS THE HOOD STRUCTURALLY SOUND? YES V NO _
2. IF THERE ARE HOLES, DENTS, ETC., WHERE ARE THEY?
C. L. DOES THE HOOD HAVE A FILTER? YES \/ NO _
2. WHEN WAS IT LAST INSPECTED? /Sl\./"Pt LAST CHANGED?
III. GENERAL OBSERVATIONS
A. ARE THE VOC EMITTED AT A NOTICEABLE RATE? YES _V/NO _
B. ARE THERE CROSS-DRAFTS IN THE ROOM? YES S NO _
C. IS THE HOOD WELL POSITIONED TO CAPTURE THE VOC? YES V NO
D. WHAT IS THE APPROXIMATE DISTANCE FROM THE EMISSION POINT TO
THE HOOD OPENING? 3 FT O INCHES
E. 1. DOES THE HOOD APPEAR TO HAVE BEEN MODIFIED OR ALTERED IN
ANY WAY? YES NO \S
2. IF YES, HOW?
(continued)
A-2
-------
Table A.I --Continued
F. 1. DOES THE HOOD CAPTURE THE VOC? YES NO
2. IF NOT, WHAT IS HAPPENING WITH THE VOC? 'TU/ NO
2. IF SO, WHERE?
3. DOES IT, SEEM TO BE RELATED TO THE HOOD CAPTURE EFFICIENCY?
YliS V NO
H. OIJSER VAT IONS
IV. MEASUREMENTS
A. TYPE OF INSTRUMENT USED
B. WHERE WAS MEASUREMENT(S) TAKEN? (DRAW SKETCH BELOW)
C. INSTRUMENT READING(S) (IF APPLICABLE)
D. OBSERVATIONS
V. SKETCH PROCESS, LOCATION OF HOOD, POSITION OF MEASUREMENT DEVICE
VI. GENERAL COMMENTS:
A-3
-------
TABLE A.2
CHECKLIST FOR DUCTWORK
PLANT XY2" IKS PECTOR
DATE 4-/IZ I A3-
I. SYSTEM I.AYOUT
A. 1. SKETCH BELOW THE WAY IN WHICH THE DUCTWORK TIES TOGETHER THE
HOODS AND THE CONTROL EQUIPMENT.
2. LABEL INDIVIDUAL BRANCHES.
ft. HOW MANY HOODS ARE CONNECTED TO THE DUCTWORK? 3
C. 1. ARE ALL HOODS CURRENTLY IN USE? YES V NO
2. IF NO, MARK UP SKETCH TO SHOW WHICH ARE/ARE NOT CONNECTED.
I). PHYSICAL INSPECTION
1. WHICH BRANCHES DID YOU INSPECT? (LIST) FTLL
2. NOTE THE APPEARANCE OF ANY BRANCH WHICH APPEARS TO BE IN POOR
CONDITION (DENTS, RUST, HOLES, ETC.)
3. NOTE THE NUMBER AND CONDITION OF ANY BENDS, ELBOWS AND TRANSITION
PIECES WHICH ARE IN POOR CONDITION
4. DO ANY PUiCKS OF DUCTWORK APPEAR NEWER THAN OTHERS?
YES V^NO _ (IF YES, NOTE WHERE ON SKETCH)
5. WHY WAS DUCTWORK CHANGED?
6. DO ANY SECTIONS OF THE DUCTWORK APPEAR VULNERABLE TO BEING HIT
BY A MOVING CART, FORKLIFT, CRANE, ETC.? YES NO \£ (IF
7. HOW OFTEN ARE WJCTS INSPECTED FOR MATERIAL (POLYMERS, JU2SINS,
ETC.) BUILDUP ON THE INSIDE? / PER ID ftUrvdOlA^
(coQtiuued)
A-4
-------
LI.
TABLE A.2 "Continued
E.
V/ N
DAMPERS
1. ARE DAMPERS USED TO ISOLATE DUCTWORK BRANCHES? YES V NO
2. WHERK AKK THE DAMPERS LOCATED? (NOTE ON SKETCH)
3. WHAT IS EACH DAMPER'S FUNCTION? iMERGr.NCY BYPASS
FLOW CONTROL ~
OTHER (SPECIFY)
4. IS DAMPER MANUALLY OR AUTOMATICALLY ACTIVATED?
5. WHAT IS THE POSITION OF EACH DAMPER DURING THE INSPECTION?
(0 DECREES - FULL OPEN, 90 DEGREES - FULL CLOSED) - O
6. IS EACH DAMPER CLEARLY LABELED? YES NO
SKETCH OF SYSTEM
III. GENERAL COMMENTS :
A-5
-------
TABLE A.3
CHECKLIST FOR FANS
xy-2-
PLANT _
I. TECHNICAL DATA
A. FAN MOTOR
1. MANUFACTURER
II.
INSPECTOR
DATE
2. RATED HORSEPOWER 7*5"
B. DRIVE
1. DIRECT BELT
3. MAXIMUM RPM
v/
OTHER
2. PULLEY REDUCTION
C. FAN
1. MANUFACTURER
2. INSTALLATION DATE
3/76
3. RPM
OPERATING AND MAINTENANCE DATA
A. FAN MOTOR
1. NOTICEABLE OVERHEATING? YES NO
2. HOW OFTEN IS BEARING INSPECTED?
3. BEARING LAST CHANGED (DATE)
4. SPARE BEARING KEPT IN STOCK? YES
••MMiM
5. COMMENTS
B. DRIVE
1. AUDIBLE BELT SLIPPAGE? YES NO
2. BELT CONDITION
PER
NO
(continued)
A-6
-------
TABLE A.3 --Continued
3. HOW OFTEN IS BELT INSPECTED?
4. WHEN WAS BELT LAST CHANGED? f*/8I LAST ADJUSTED?
5. SPARE BELT IN STOCK? YES S NO
6. COMMENTS "^yLCl^x'^^'*'^
7
C. FAN
1. NOTICEABLE FAN VIBRATION?
2. FAN BLADE CONDITION
3. HOW OFTEN IS BEARING INSPECTED? / PER
4. BEARING LAST CHANGED
5. SPARE BEARING KEPT IN STOCK? YES r NO _
6. DAMPER INSTALLED AT FAN INLET? YES _ NO Y
7. DAMPER POSITION Z OPEN
8. FAN STATIC PRESSURES:
: INLET "7>Wo OUTLET 3 P±*-
9. CONTINUOUSLY MEASURED VARIABLES?
a.
b.
c.
FAN SPEED
AIR FLOW RATE
INLET STATIC
PRESSURE
YLS NO X
YES ^ NO
YES X NO
d. OUTLET STATIC
PRESSURE YES V NO
III. GENERAL COMMENTS:
A-7
-------
TABLE A.4
INCINERATOR INSPECTION CHECKLIST
(For each incinerator, fill out a separate checklist.)
COMPANY:
WCATION:
INSP. DATE:
INSPECTOR:
ID:
I. EQUIPMENT IDENTIFICATION
A. Control Device ID:_
Model;
Manuf:
B. Date Installed:,
C. Pcnit No.ls)
IX. VOC EMISSION SOURCES VENTED TO CONTROL DEVICE;
Process ID
Process Sources
III. CONTROt DEVICE DESCRIPTION
Incinerator
(1) Type Incinerator (Choose more than one if appropriate.)
__ Thermal ___ Single Chamber __^ Controlled Air
Catalytic Mult. Chamber- Otnar (Specify)
(2) Burner Fuelt
____ Natural Gas
___ 011» Sulfur Content % by Height
Other (Specify) ; Sulfur Content % by Height
(3) Catalyst (If applicable)
Types
Heat Recovery
Is heat recovered?
Yes
No
A-8
-------
TABLE A04 (cnt'd)
Page 2
Ventilation System Fan/Motor
(1) Type Fan;
Axial
_____ Centrifugal
Other (Specify)
(2) Type Motor:
Belt Driven
_____ Direct Drive
Other (Specify)
(3) Location with Respect
15 Control Device;
_____ Upstreaa
Downstream
Emergency Shutdown
Reason
Parameter
Monitored
location of
Monitor
Is Shutdown
Automatic?
tY/N)
Does It
Shutdown the
Process?
(Y/N)
III. OPERATIONAL MONITORS
Parameters
Incinerator
• Inlet Teap.
• Outlet Teap.
e Combustion
Chamber leap.
• Inlet VOC
e Outlet VOC
e Burner fuel Usage
e Catalyst Teap.
e Outlet Flow Rate
e
e
e
Vent. Fan/Motor
e RPM
e Pressure Drop
e Across Fan
Monitored?
(Y/N)
Continuous
Monitor?
(Y/N)
.
Recorder?
(Y/N)
-
Reading
Units
A-9
-------
TABLE A.5
COMPANY:
LOCATION:_
ID NO.:
ADSORBER INSPECTION CHECKLIST
(For each adsorber, fill out a separate checklist.)
INSP. DATE:_
INSPECTOR:
I. EQUIPMENT IDENTIFICATION
A. Control Device ID:_
Model:
Manuf:
B. Date Installed:,
C. Perait No.(s):
II. VOC EMISSION SOURCES VENT TO CONTROL^DEVICE
Process ID
Process
Sources
III. CONTROL DEVICE DESCRIPTION
A. Adsorber
(1) Type Adsorber
Fixed Bed _
Pluldlzed Bed __
__^_ Continuous
(2) Adsorbent
Type; _^__ Throwaway
___ Activated Carbon __
__— Renewable
Other (Specify)
(3) Used in service with other control devices?
If "Yes," specify:
Concentric
Other (Specify)
Fora:
Granular
Pelletlzed
Powder
Yes
No
A-10
-------
B.
Ventilation System Fan/Motor
(1) Type Fan (2) Type Motor
Axial Belt Drive
_____ _____ Direct Drive
Other (Specify)
Page 2
(3) Location with Respect
to~Control Device ""
______ Upstream
Downstream
C.
Regeneration Systea
(1) Method
Steaa
• Pressure:
psla
Hot Air Inert Gas
• Temperature: T • Type gas:
(2) Initiation Method
_____ Concentration
• Concentration Set Point:
• Concentration Monitor
Location;
(3) Baisslon Controls
Type;
Condensor/Decantor
Distillation
Other (Specify)
Timed
• Time between Regeneration; hrs.
• Time for Regeneration: mlns.
Handling of Removed Material;
Reclaimed/Revised On-slte
Hauled Off-site
0. Emergency Shutdown
Reason
Parameter
Monitors
Location of
Monitor
Is Shutdown
Automatic?
(T/N)
Docs It
Shutdown the
Processes?
(Y/N)
A-ll
-------
III. OPERATIONAL MONITORS
Paraaeters
Adsorber
• Pressure Drop
Across Bed
• Pressure Drop
Across Premiers
• Vapor Inlet Tenp.
• Bed Tenp.
• Vapor Outlet Tenp.
• Inlet L£L
• Inlet VOC Content
• Outlet VOC Content
•
•
•
•
Regenerator
• Steam Pressure
• Tenperature
•
•
•
Fan /Motor
• RPM
• Pressure Drop
Across Paa
Monitored
(Y/N)
Continuous
Monitor
(Y/N)
•
Recorder
(Y/N)
Reading
Units
A-12
------- |