EPA 340/1-77-019
OCTOBER 1977
Stationary Source Enforcement Series
INSPECTION MANUAL FOR ENFORCEMENT OF
NEW SOURCE PERFORMANCE STANDARDS
FUEL GAS FIRED COMBUSTION UNITS
(REFINERIES)
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Enforcement
Office of General Enforcement
Washington, D.C. 20460
-------
INSPECTION MANUAL FOR THE
ENFORCEMENT OF NEW SOURCE
PERFORMANCE STANDARDS:
FUEL GAS FIRED COMBUSTION UNITS
Prepared by
George E. Umlaut and Bansi Parekh
Contract No. 68-01-3156
EPA Project Officer
Mark Antell
Prepared for
UNITED STATES
ENVIRONMENTAL PROTECTION AGENCY
Division of Stationary Source Enforcement
Washington, D.C.
August 1976
-------
This report was furnished to the United States Environmental
Protection Agency by The.Ben Holt Co., Pasadena, California,
in fulfillment of Contract No. 68-02-1090 and by Pacific
Environmental Services, Inc., Santa Monica, California in ful-
fillment of Contract No. 68-01-3156. The contents of this
report are reproduced herein as received from the contractor.
The opinions, findings, and conclusions expressed are those
of the author and not necessarily those of the Environmental
Protection Agency.
-------
TABLE OF CONTENTS
Page
LIST OF FIGURES vi
1.0 INTRODUCTION 1-1
2.0 STATE IMPLEMENTATION PLANS (SIP) AND NEW
SOURCE PERFORMANCE STANDARDS (NSPS) 2-1
2.1 Existing Sources - Typical SIP Requirements 2-1
2.1.1 Summary of Typical SIP Requirements 2-1
2.2 New Source Performance Standard (NSPS) Requirements 2-2
2.2.1 Emission Standards 2-2
2.2.1.1 Exemptions 2-3
2.2.2 Performance Testing 2-3
2.2.2.1 Initial Performance Test 2-3
2.2.2.2 Subsequent Performance Tests 2-3
2.2.3 Monitoring Requirements 2-4
2.2.4 Record Keeping and Recording 2-4
2.2.4.1 Notifications Regarding Initial
Start-Up 2-4
2.2.4.2 Start-Up, Shut-Down, and
Malfunction Recording 2-5
2.2.4.3 Records Regarding Performance
Testing 2-5
2.2.4.4 Quarterly Reports 2-5
2.3 Applicability of Standards 2-6
2.3.1 Sulfur Dioxide Emission Standard 2-6
m
-------
TABLE OF CONTENTS (continued)
Page
3.0 PROCESS DESCRIPTION, ATMOSPHERIC EMISSIONS
AND EMISSION CONTROL METHODS 3-1
3.1 Fuel Gas Combustion System 3-1
3.1.1 Acid Gas Treatment Plant 3-1
3.1.2 Fuel Combustion Units 3-4
3.2 Atmospheric Emissions 3-4
3.3 Pollution Control Methods 3-6
3.4 Process Instrumentation 3-6
4.0 START-UP/MALFUNCTIONS/SHUT-DOWN 4-1
4.1 Start-up 4-1
4.2 Malfunction 4-1
4.3" Shut-Downs 4-2
5.0 INSPECTION PROCEDURES 5-1
5.1 Conduct of Inspection 5-1
5.1.1 Formal Procedure 5-1
5.1.2 Overall Inspection Process 5-2
5.1.3 Safety Equipment and Procedures 5-3
5.1.4 Frequency of Inspections 5-3
5.2 Inspection Checklist 5-4
5.2.1 Facility Identification 5-4
5.2.2 Opacity Observations 5-5
5.2.3 Odors 5-5
5.2.4 Acid Gas Treating Plant 5-5
5.2.5 Fuel Gas Combustion Unit 5-7
5.2.6 Air Pollution Control Equipment 5-8
5.2.7 Records of Operation 5-9
5.2.8 Operational Aspects 5-9
5.3 Inspection Follow-Up Procedures 5-1Q
-------
TABLE OF CONTENTS (continued)
Page
6.0 PERFORMANCE TEST 6-1
6.1 Process Operating Conditions 6-1
6.2 Process Observations 6-2
6.2.1 Acid Gas Treatment Plant 6-2
6.2.2 Fuel Gas Combustion Device 6-3
6.3 Emission Test Observations 6-3
6.3.1 Traversing 6-3
6.3.2 Stack or Duct Gas Velocity
Determination 6-4
6.3.3 Gas Analysis ' 6-5
6.3.4 Moisture 6-6
6.3.5 Hydrogen Sulfide 6-6
6.3.6 Sulfur Dioxide 6-8
6.4 Emissions Monitoring 6-9
6.4.1 H2S Monitoring 6-9
6.4.2 S02 Monitoring 6-11
-------
LIST OF FIGURES
Figure Page
3-1 Acid Gas Treating Plant 3-3
3-2 A Large Box-Type Refinery Heater 3-5
3-3 A Vertical, Cylindrical Refinery Heater 3-5
VI
-------
1.0 INTRODUCTION
Pursuant to Section 111 of the Clean Air Act (42 USC 1857
et. seq.)> the Administrator of the Environmental Protection
Agency (EPA) promulgated sulfur in fuel standards for the burning
of fuel gas in a new or modified fuel combustion device in a re-
finery. These proposed standards were issued in the Federal
Register of June 11, 1973, and final standards (40 CFR 60.104)
became effective on February 28, 1974. The standards apply to
all sources whose construction or modification commenced after
June 11, 1973. A new emission source is one which is designed
and constructed after the formal proposal of new source regula-
tions. New sources include newly constructed facilities, new
equipment which is added to existing facilities, and existing
equipment which is modified in such a way that results in an in-
crease of pollutant emissions. New source standards limit spe-
cific pollutant emissions from categories of sources (such as
fossil fuel-fired steam generators, municipal incinerators)
which the Administrator determines may contribute significantly
to the endangerment of public health and welfare. For these
sources, the Act requires the Administrator to promulgate emis-
sion limitations which will require installation of the best sys-
tems of emission reduction which he determines have been ade-
quately demonstrated. Cost factors are considered in making this
determination. Federal new source standards help prevent the
occurrence of new air pollution problems, encourage improvements
in emission control technology, and provide a mechanism for con-
trolling pollutants which EPA suspects are hazardous, but for
which insufficient information is available to regulate such
pollutants under other provisions of the Act.
The intent of this report is to provide guidelines for the
appropriate enforcement agency in the development of inspection
programs for fuel gas combustion systems at a refinery which are
covered by New Source Performance Standards (NSPS). Section 2.0
presents the NSP standards as well as a brief outline of related
State Implementation Plan (SIP) regulations. Process descrip-
tions for fuel gas treatment plants and fuel combustion devices
along with the associated emissions and control techniques are
summarized in Section 3.0. Section 4.0 discusses the start-up,
shut-down as well as malfunction problems associated with the
fuel gas treatment process and fuel gas combustion devices. De-
tailed inspection procedures required by appropriate personnel
during an inspection are presented in Section 5.0. Discussion
of performance test requirements is presented in Section 6.0.
1-1
-------
2.0 STATE IMPLEMENTATION PLANS (SIP) AND
NEW SOURCE PERFORMANCE STANDARDS (NSPS)
This section briefly summarizes state implementation plan
regulations and NSPS for the emission of sulfur dioxide from fuel
combustion equipment. State implementation plans are federally
approved and are designed to prevent local ambient air concentra-
tions from exceeding the National Ambient Air Quality Standards.
The NSP standards apply to fuel gas combustion devices whose con-
struction or modification commenced after June 11, 1973.
2.1 EXISTING SOURCES - TYPICAL SIP REQUIREMENTS
A nationwide review of the state implementation plan regula-
tions for the emissions of sulfur oxide from fuel combustion
has been recently published by the Strategies and Air Standards
Division of EPA, 1 Research Triangle Park. This summary of SIP
emission requirements shows the complexity and diversity of S02
emission regulations for fuel combustion. Some states control
all emission sources equally, while other states prescribe differ-
ent emission limits for sources according to the fuel used, the
geographic location, the size of the source, or the type of
source. Following sections highlight typical emission limita-
tions.
2.1.1 Summary of Typical SIP Regulations
A review of Reference 1 reveals that S0« emissions are
most commonly regulated by limiting the amount of sulfur or sul-
fur dioxide emitted per unit heat input (LB S/MMBTU, LB S02/MMBTU)
or by limiting the sulfur content by weight percent that a fuel
can contain. S02 limitations are also expressed in parts of S02
per million parts by volume of stack gas or the weight of S02
emitted per hour (LB S02/HR). Other methods of limiting S02
emissions which appear in the SIPs include requiring a percent
control of input sulfur or requiring application of "latest
reasonably available control technology" (Florida) or "new
proven technologies" (Texas). A few states have SOX emission
limitations addressed directly to petroleum refinery operations
or written in terms of H2S concentrations in refinery process gas
streams.
2-1
-------
Typical examples of currently approved SOX regulations for
combustion units are:
• The California South Coast Air Quality Management
District prohibits the burning of any gaseous fuel
containing sulfur compounds in excess of 800 ppm (50
grains per 100 cubic feet) of gaseous fuel, calculated
as hydrogen sulfide at standard conditions, or any
liquid fuel or solid fuel having a sulfur content in
excess of 0.5 percent by weight (Rule 431).
t The Jefferson County, Kentucky, Air Pollution Control District
prohibits combustion of any refinery process gas stream that
contains H2$ in concentration greater than 10 grains per 100
cubic feet of gas without removal of the H2S in excess of this
concentration (Rule 4.2.1).
Most of the state implementation plans were approved in 1972.
Following initial approval of the SIPs, many states began submit-
ting to EPA revisions to their implementation plan, many of which
alter the emission limitations. Usually, these revisions are
based on additional air quality measurement data or on a more de-
tailed technical analysis of air pollution control strategies.
When approved by EPA, these revisions become a part of the imple-
mentation plan.
2.2 NEW SOURCE PERFORMANCE STANDARD (NSPS) REQUIREMENTS
The following sections describe in detail the requirements
outlined in NSPS for new refinery process gas combustion units.
The standards are given in terms of emissions, performance testing,
monitoring, record keeping and reporting.
2.2.1 Emission Standards
The NSPS regulations for S02 emissions from refinery combus-
tion units appeared in 40 CFR 60.104(a) and state that, "No owner
or operator subject to the provisions of this subpart shall burn
in any fuel gas combustion device any fuel gas which contains
H2S in excess of 230 mg/dscm (0.1 gr/dscf)."2 An alternate pro-
vision (40 CFR 60.104(b)) to this regulation states, that, "The
owner or operator may elect to treat the gases resulting from the
combustion of fuel gas in a manner which limits the release of
S02 to the atmosphere if it is shown to the satisfaction of the
Administrator that this prevents S02 emissions as effectively as
compliance with the requirements of paragraph (a) of this sec-
tion. "2 In other words, the regulation requires treating of fuel
gas or combustion products for any new fuel gas combustion de-
vice where the H2S content of the fuel gas is greater than 230
mg/dscm (0.1 gr/dscf).
2-2
-------
2.2.1.1 Exemptions
The proposed standard does not apply to unusual situations,
such as emergency gas releases. As stated in the regulation, the
combustion of process upset gas in a flare, or the combustion in
a flare of process gas or fuel gas which is released to the flare
as a result of relief valve leakage is exempted from the require-
ments of the regulation as stated in Section 2.2.1. However,
flares which are maintained to burn a continuous process gas
stream are subject to the requirements of the NSPS regulations.
2.2.2 Performance Testing
The owner or operator of a new fuel gas burning device is
required to conduct performance tests within a specified period
after start-up, and thereafter from time to time as may be speci-
fied by the EPA. These performance tests are required in order
to demonstrate that the standards are being met by the new de-
vice.
2.2.2.1 Initial Performance Test
The initial test of performance of a new facility must be
conducted within 60 days after the facility is first operated
at its maximum intended rate of operation. However, if the in-
tended rate of operation is not achieved within 120 days of
initial start-up, the initial test must nevertheless be conducted
within 180 days of start-up. Thirty days must be allowed for prior
notice to the EPA, to allow the Agency to designate an observer
to witness the test.
Each performance test must be conducted in accordance with
the instructions set forth in the regulations (Reference 3),
which are discussed in more detail in Section 6 of this manual.
The test consists of three repetitions of the specified procedure;
performance of the facility is judged acceptable if, for each of
the characteristics tested, the average value from the three
repetitions is less than the NSP Standard value.
Necessary modifications in the details of the test methods
may be made, if approved in advance by the EPA. A written re-
port of the test is to be furnished to the EPA.
2.2.2.2 Subsequent Performance Tests
Subsequent to the initial test, further performance tests
may be required from the source at the discretion of the EPA.
Alternatively, .the Agency may decide to conduct performance tests,
for which purpose the owner or operator is required to provide
2-3
-------
testing facilities, which include necessary utilities, sampling
platforms and safe access to the sampling platform.
Performance testing subsequent to the initial test is most
likely to be required when records indicate a relatively high
frequency of occurrence of emission levels near, at, or above
the NSPS levels.
2.2.3 Monitoring Requirements
The NSP Standards (40 CFR 60.105) require that the owner
or operator of any fuel gas burning device in a refinery subject
to the provisions of the regulation as discussed in Section 2.2.1
shall install, calibrate, maintain, and operate an instrument
for continuously monitoring and recording concentrations of H2S
in fuel gases burned in any fuel gas combustion device, or an
instrument for continuously monitoring and recording concentra-
tions of S02 in the gases discharged into the atmosphere from
the combustion of fuel gases when the owner or operator elects
to treat the gases. It should be noted, however, that no HpS
emission monitoring method has as yet been specified in the
regulations. Until acceptable monitoring apparatus have been
approved, this monitoring requirement can not be enforced.
When a refinery has several fuel gas combustion devices
having a common source of fuel gas, monitoring may be done at
one location if sampling at this location produces results repre-
sentative of the H2S concentration in the fuel to all units.
This situation is common in many refineries where a centralized
acid gas treatment plant is maintained which treats HoS rich
gases from several refinery process units and routes the treated
exit gases to refinery combustion unit fuel supplies.
2.2.4 Record Keeping and Reporting
Any owner or operator subject to the requirements of NSPS
is required to maintain records, to furnish certain reports,
and to notify EPA of certain plans and occurrences, as discussed
below [40 CFR 60.7].
2.2.4.1 Notifications Regarding Commencement of Construction
The owner or operator of any facility subject to the pro-
visions described in Section 2.3 of this report shall submit the
Administrator of the EPA a notification of the date of commence-
ment of construction of the affected facility. This notification
must be postmarked no later than thirty (30) days after such date.
2-4
-------
2.2.4.2 Notifications Regarding Initial Start-Up
The owner or operator is required to furnish to the Adminis-
trator of the Environmental Protection Agency written notifications
of anticipated initial start-up date and actual start-up date of
the new facility. The notification of the aniticpated start-up
date must be postmarked no more than sixty (60) days nor less
than thirty (30) days prior to that date; notification of actual
start-up must be postmarked within fifteen (15) days after its
occurrence. In this connection, "start-up" refers to operation
of the facility for any purpose,
2.2.4.3 Start-up, Shut-down, and Malfunction Recording
The owner or operator is required to record the occurance
and duration of any start-up, shut-down, or malfunction in oper-
ation of the fuel gas burning device and to retain the record
for at least two years thereafter. The record should also
include the nature and cause of any malfunction, together with
a notation as to the corrective action and any measures taken
to prevent recurrence of the malfunction.
In this connection, "start-up" refers to a renewed opera-
tion of the facility for any purpose; "shut-down" means the
cessation of operation of the facility for any purpose; and "mal-
function" is defined as any sudden unavoidable failure for any
component of the plant or of the fuel gas burning device itself
to operate in a normal manner. Preventable failures, such as
those which may have been caused by poor maintenance or careless
operation, or by equipment breakdown due to such causes, are not
included in this definition.
2.2.4.4 Records Regarding Performance Testing
The owner or operator is required to make available to the
EPA, in order to facilitate conduct of performance tests by the
Agency, any records necessary to determine whether performance
of the fuel gas combustion is a representative performance at
the time of the test. Fuel gas burn rate and hours of operation
for any fuel gas combustion device shall be recorded daily. A
file of all measurements required for H2S emissions regulations
shall be maintained by the owner or operator. Appropriate meas-
urements shall be reduced if necessary to the units of applicable
standard daily, and summarized monthly. The record of any such
measurement(s) and summary shall be retained for at least 2
years following the date of such measurements and summaries.
These records should also be made available prior to or during
inspection of the facility.
2-5
-------
2.2.4.5 Quarterly Reports
Quarterly reports are to be filed on the 15th day following
the end of each calendar quarter. These reports must include the
records of excessive emissions during the calendar quarter in
terms of date, time of commencement, and time of completion of
each period of excessive emissions, as evidenced by records of
monitoring equipment or other observations. The quarterly re-
ports must also include the records of start-up, shut-down, and
malfunction during the calendar quarter, with details as to the
causes of malfunctions and corrective measures applied.
2.3 APPLICABILITY OF STANDARDS
The NSP Standards are applicable to all new or modified fuel
gas combustion devices including process heaters, boilers and
flares used to combust fuel gas. A new or modified boiler which
has a heat input greater than 250 million BTU per hour may also be
subject to the provisions of 40 CFR Section 60, Subpart D. A new
or modified source is one on which construction or modification
commenced after June 11, 1973, and subject to the following defini-
tions (40 CFR § 60.2).
• "Modification" means any physical change in, or change in
the methods of operation of, an affected facility which
increases the amount of any air pollutant (to which a
standard applies) emitted by such facility or which re-
sults in the emission of any air pollutant (to which a
standard applies) not previously emitted. The following shall
not, by themselves be considered to be modifications.
(1) Routine maintenance, repair and replacement
(2) An increase in production rate of an existing facility
if that increase can be accomplished without a capital
expenditure on the stationary source containing the
facility
(3) An increase in hours of operation
(4) Use of alternative fuel or raw material if, prior to
June 11, 1973, the existing facility was designed to
accommodate such alternate use. Conversion to coal
required for energy considerations ( §119 (d) (5) ) shall
not be considered as a modification.
(5) The addition or use of any system or device whose primary
function is the reduction of air pollutants, except when
an emission control system is removed or replaced by a
system which the Administrator determines to be less
environmentally beneficial.
(6) The relocation or change in ownership of an existing facility
2-6
-------
• "Commenced" means that an owner or operator has under-
taken a continuous program of construction or modification
or that an owner or operator has entered into a binding
agreement or contractual obligation to undertake and com-
plete, within a reasonable time, a continuous program of
construction or modification.
2.3.1 Sulfur Dioxide Emission Standard
The performance standard for emission of sulfur dioxide,
applicable to new fuel gas combustion devices, is written in
terms of H2S concentration of fuel gas. This concentration of
H2S in fuel gas is limited to 230 milligrams per dry standard
cubic meter (mg/dscm) of fuel gas, or, in English units, 0,1
grains per dry standard cubic foot (gr/dscf). No owner or oper-
ator is permitted to burn any fuel gas containing more than the
above specified concentration of ^S in any fuel gas combustion
unit. Actual H2S concentration of fuel gas must be monitored
during routine operation of facility. A determination as to whether
the facility is in or out of compliance with this regulation is
based on the results of performance tests conducted in the manner
prescribed in 40 CFR § 60.106 (c).
Since a refinery generally operates several fuel gas com-
bustion units, performance tests must be conducted on the fuel
gas going to each applicable combustion device. However, if a
refinery has a centralized fuel gas supply system with an acid
gas treatment plant for all of the fuel gas streams, then a one
point performance test of H2S concentration at the outlet of fuel
gas supply system can be performed to determine compliance for all
of the affected combustion units.
The sampling period for the performance test is specified in
40 CFR § 60.106 (c)2 and is not less than 10 minutes. Since three
repetitions are required the total minimum sampling time is 30
minutes.
As discussed in Section 2.2.1 the second provision of the
S02 regulation [40 CFR § 60.104 (b)]2 may allow burning of fuel
gas with an H2S concentration greater than 0.1 gr/dscf provided
that the owner or operator elects to treat the combustion gases
so as to reduce the S02 emissions to atmosphere as effectively
as compliance with the requirements of 40 CFR § 60.104 (a).2 in
order to determine whether the facility is or is not in compliance
with the regulation, a performance test should be conducted at the
emission point as described in 40 CFR § 60.106 (d).2 Detailed
discussion of this test method will be covered in Section 6,0.
2-7
-------
3.0 PROCESS DESCRIPTION, ATMOSPHERIC EMISSIONS
AND EMISSION CONTROL METHODS
A refinery, in general, is a large industrial complex consist-
ing of many processes which require fuel combustion units for
process heat. Many of the refinery processes produce gas streams
which can be used as fuel in combustion units. These gases general-
ly contain a variety of sulfur compounds which in turn can be
oxidized to SOg and be emitted to the atmosphere from combustion
units. The amounts and types of the sulfur compounds in the fuel
gases from a unit depend upon the sulfur content of the feedstock
to the unit and the process operating parameters such as temperature,
pressure or catalyst type.
In order to reduce the sulfur content of these acid gas
streams and make them more suitable for fuel usage, refineries
utilize amine absorption processes. Normally, one HDS
unit will be maintained to treat a number of process gas streams
generated at different refinery units and will provide cleaned fuel
gas for a number of combustion units. The scope of this section
will be to consider the operating conditions, atmospheric emission
points and possible emission control systems starting at the inlet
to the acid gas treating facility. While the NSPS standards are
only aimed at control of new refinery combustion units, it is the
operation of the acid gas treater which actually determines what
the sulfur content of the fuel gas will be and hence the compli-
ance of the fuel burning source.
3.1 FUEL GAS COMBUSTION SYSTEM
As discussed in the proceeding section several processes such
as cracking, coking, hydrocracking and other fractionation units
are responsible for the production of refinery fuel gases. These
gases may be treated and then used as fuel in combustion units or
may be used directly in combustion units.
3.1.1 Acid Gas Treatment Plant
The Acid Gas Treating process uses either diethanolamine CDEA)
or monoethanolamine (MEA) in aqueous solution to remove acid gases
(HzS and C02) from the process gas stream. Removal is accomplished
by a chemical reaction between the acid gases and the chemically
basic amine solution. The reaction can be reversed by heating the
solution, and this behavior is used to regenerate the treating
solution.
3-1
-------
A process flow diagram for a typical Acid Gas Treating Plant
is shown in Figure 3-1. The process gas containing H2S (sour gas),
is fed to the bottom of the amine absorber where it is brought into
counter-current contact with regenerated (lean) amine solution.
The treated gas leaves the top of the absorber and is sent to the
refinery fuel gas system. The amine-acid gas (rich amine) leaves
the bottom of the absorber, is heated by exchange with stripper
bottoms, and is sent to the top of the stripper, Acid gas is
stripped from the amfne by steam generated at the bottom of the
stripper by the stripper reboiler. The acid gas and steam leave
the top of the stripper, pass through a condenser and enter the
separator where the condensed steam is separated from the acid gas.
Acid gas from the separator is sent to a sulfur plant where the
H2S is converted to elemental sulfur. The condensed steam is re-
turned to the stripper as a reflux. Regenerated Clean) amine from
the bottom of the stripper is cooled by exchange with rich amine
and is pumped through an amine cooler back to the top of the ab-
sorber. Part of the lean amine is sent through an amine filter to
maintain solution cleanliness and a sump is provided for adding or
draining solution to- or from the system. Further information on
acid gas treating can be obtained from References 4 and 5.
The degree of treatment achieved depends mainly on the par-
tial pressure of H2S in the treated gas in equilibrium with the
incoming lean amine. This in turn is a function of operating
pressure, the residual H2$ content of the lean amine, the amine
concentration, and the temperature. Higher operating pressure
results in higher removal efficiency but operating pressure is
normally fixed by the pressure required in the fuel gas system,
usually about 3.52 kg/cm^ (50 psig), and the cost of compressing
process gas. The H2S residual concentration in the lean amine is
set by stripper operating conditions. Low pressure favors good
stripping and the pressure is generally held at a few psig to sup-
ply the sulfur plant without the use of compressors. Increasing
heat to the stripper reboiler provides more stripping steam and
better stripping, but the amount of steam is limited by cost and
equipment size. Amine concentration must be maintained within
limits and should be checked regularly. Too high an amine con-
centration leads to excessive stripper bottoms temperature, with
corrosion and amine degredation. Low amine concentration can be
partially compensated for by increased amine circulation, within
equipment limitations, but this increases operating costs. Low
amine temperature favors high treating efficiency, and proper
cooling of the lean amine depends on adequate sizing and main-
tenance of the rich amine--lean amine heat exchanger and the amine
cooler. The amine cooler may be a fin-fan unit, in which case
performance is affected by air temperature, or a cooling water
exchanger, in which case cooling water temperature is important,
3-2
-------
CO
CO
THE BEN HOLT CO.
-------
3.1,2 Fuel Combustion Units
In the context of this report, an affected fuel combustion
unit has been deffned at the onset of Section 2.3. A general
discussion will be presented to furnish the reader with more in-
formation on the types of combustion units used in a refinery and
the types of fuels and associated emissions which are.typical.
Refinery combustion operations involve the use of process heaters
to transfer heat to hydrocarbon fluids in heat exchange devices
as well as the use of boilers to produce process steam and flares
to burn waste gases. Process heaters and boilers are fired by
either natural gas, refinery process gas, oil, or a combination
of these fuels. Gaseous fuels used are usually mixtures of non-
condensible hydrocarbons that for one reason or another are not
marketed or further processed.
Figures 3-2 and 3-3 show the types of heaters used typically
in a refinery. In most refinery heaters an oil or other petroleum
product flows inside the heat exchange tubes. Boilers and re-
boilers are generally used for the process steam requirements or
may be used for power generation. These boilers are either gas,
oil and/or coal fired. Use of a particular type of fuel will
depend on the availability of the fuel as well as air pollution
regulation constraints.
Treated refinery gas leaves the top of the absorber of the
acid gas treatment plant and is collected in a fuel gas system
as discussed earlier in Section 3.1.1. Treated gas from this fuel
gas system is distributed to various combustion units according to
their requirements. As an alternative, a refinery may elect to
use the refinery gas as it is, i.e. without acid gas treatment,
and treat the exhaust gases for reducing the pollutants emitted.
3.2 ATMOSPHERIC EMISSIONS
Except for leaks and malfunctions, the treating plant is
not in itself a source of atmospheric emissions. If, however, the
plant is not operating efficiently, the treated gas can contain
excessive amounts of ^S and burning the gas can lead to excessive
sulfur dioxide emissions to the atmosphere, Also, the acid gas
product of the plant has high H2S concentrations and must be
properly disposed of, usually in a sulfur plant.
Combustion units are the ultimate emission source of this
system. Most of the sulfur contained in the fuel to the unit is
oxidized to S02 and leaves the system as combustion gases. Com-
bustion units can also be sources of other emissions which are not
3-4
-------
•—•••» ^.
•: : .• » -
3
0
O
c
0
c
0
c
o
o
c
s
3_
h
0 3
o
0 t
c
O 0
o
3
^1'
1 J -"
•> o
0 0 C
o o
o o c
,^
*-—
-1 •
•J
0 '
0
0 0
u - o .j . : Ji .? .. . y
.-
0
c
o"
•3
o
o
o
o
0
0
c
1, _c
n
Figure 3-2 A large box-type refinery heater.
TO STICK
MR
•• ••«
IN
IIR PREHEtT
,OOOO
O O O O O O
'O O O O O
|o o c o o ol
00,00
Figure 3-3 A vertical, cylindrical refinery heater
3-5
-------
regulated by NSPS. They can emit both visible and non-visible air
contaminants. Visible contaminants are principally liquid and
solid particulates. Non-visible contaminants include oxides of
nitrogen and carbon. Inorganic ash content of oils and coal after
nearly complete combustion of these fuels constitutes the partic-
ulate emissions. At times when soot blowers are in operation,
particulate matter concentrations in exit gases increase markedly.
Emissions of oxides of nitrogen from combustion equipment result
from fixation of atmospheric nitrogen in the fire boxes. Com-
bustion equipment has been traditionally associated with visible
smoke plumes caused by unburned carbon and organics.
3.3 POLLUTION CONTROL METHODS
The acid gas treatment plant itself can, in a sense, be con-
sidered as a control process for the removal of H2$ from the
refinery gases before they are burned. The stripped acid gases
can be sent to a sulfur plant where h^S is converted to elemental
sulfur. Several process technologies are available for fuel oil,
gas, and coal desulfurization in order to remove the sulfur from
these fuels. When untreated fuels are used directly in combustion
units, a wet scrubber system or some sort of concentration/neutral-
ization system in conjunction with a sulfur plant can be utilized
as after gas treatment in order to reduce S02 emissions. Cyclones
and electrostatic precipitators can be used to remove particulate
matter from the exhaust gases of the combustion units when coal is
the major fuel used. A proper control of excessive air require-
ment may reduce NOX emissions to a degree.
3.4 PROCESS INSTRUMENTATION
Although there may be variations from one unit to another,
the process instrumentation shown in Figure 1 is typical for an
acid gas treatment plant. A flow recorder measures total sour
gas flow to the absorber and a pressure recorder-controller on
the treated gas maintains absorber pressure. Treated gas quality
is monitored by an ^S recorder. Amine circulation rate is set
and maintained by a flow recorder-controller on lean amine to the
absorber and rich amine leaves the absorber under level control.
Since amine circulates on a closed loop between the absorber and
stripper, the level indicator in the bottom of the stripper is an
indication of the amine inventory in the system. Reflux rate to
the stripper is set by flow control of the heating medium to the
stripper reboiler, with all condensed reflux returned to the strip-
per under level control on the separator. A recorder in the
return line records reflux flow rate. Although not shown, temper-
ature indicators are provided at key points throughout the plant
3-6
-------
including the lean amine fuel to the absorber.
Process instrumentation for a combustion device will depend
upon the type of unit, the use of the unit, and the type of fuel
used. Parameters generally controlled are fuel input rate, fuel
to air ratio, excess air requirement, fire-box temperature, and
flow rate of the fluid being heated.
3-7
-------
4.0 START-UP/MALFUNCTIONS/SHUT-DOWN
4.1 START-UP
Acid gas treating plants are sufficiently reliable in their
operation that they can be scheduled for maintenance turnaround
when the units they serve are scheduled for a turnaround. This
generally results in a plant start-up once in a period of one to
two years. The treating plant is normally brought on-stream ahead
of the units it serves so that start-up does not produce any ad-
ditional emissions over a normal operation.
Start-up procedure for a fuel combustion device depends upon
the type of combustion equipment used. Start-up procedure will
vary from unit to unit, but, in general, emissions are higher dur-
ing start-up than during normal operations. However, this excessive
emission is exempted from regulation as long as reporting require-
ments are met. The owner or operator subject to the provisions of
NSP Standards is required to furnish the Administrator written
notification of the anticipated date of initial start-up of an
affected facility not more than 60 days or less than 30 days prior
to such date. The owner or operator is also required to furnish a
notification of the actual date of initial start-up of an affected
facility within 15 days prior to such date CReference 6).
4.2 MALFUNCTION
Fire and equipment ruptures are considered to be emergencies
and first consideration must be given to safety. They are, by
their nature, unpredictable but occur rarely and their contribution
to long term polution problems is not great. For the acid gas
treatment plant power failures can result in plant shut-down, loss
of amine circulation, loss of heating medium, loss of cooling, or
loss of control system operation. Since plants are designed to fail
safe, the usual result is the flaring of sour process gas and some
acid gas product. If the power failure lasts only a few minutes,
as is often the case, it may be possible to return the plant to full
operation in a matter of minutes. Power outages of longer duration
generally result in plant shut-down. The frequency and duration of
power failures varies greatly from region to region, and may vary
with season.
Malfunctions within the process are usually related to a
failure in the supply of a sufficient quantity of cool, well-stripped,
lean amine to the absorber. The lean amine flow rate is set by the
operator to match the acid gas rate in the sour gas feed. If either
the flow rate of the sour gas or the concentration of acid gas is
4-1
-------
increased, the lean amine rate should be increased.
Types of malfunction in combustion units generally include:
• Fuel to air ratio variation
• Excessive air supply
• Fire-box refractory damage
• Erosion or corrosion of tubes in combustion units
The owner or operator is required to maintain for a period
of 2 years a record of the occurence of any malfunction that may
occur in operation of the facility (Reference 6),
4.3 SHUT-DOWNS
Scheduled shut-downs should not result in the production of
additional emissions over normal operation. By continuing strip-
per operation for a period of time after stopping the process gas
feed, the Acid Gas Treating plant can be emptied of acid gas, A
small amount of acid gas may remain in the lines to the sulfur
plant and this should be vented to a flare.
Neither should scheduled shut-downs of combustion units
result in the production ,of additional emissions over normal op-
eration. However, any emissions which occur during emergency
shut-downs are generally permitted by regulation. The owner or
operator is required to maintain for a period of 2 years, record
of the occurence of any. shut-downs that may occur (Reference 6),
4-2
-------
5.0 INSPECTION PROCEDURES
An air pollution inspection consists of entering a refinery
to determine if the equipment or processes meet the standard and
comply with the rules and regulations of the air pollution con-
trol agency. The inspection process also includes a spot-check
of selected records maintained by the operator. The enforcement
official must observe, in a qualitative manner, the items associ-
ated with atmospheric emissions. The condition and type of
equipment, and general housekeeping all influence the emission
rate. Equipment design is a major factor that must be reviewed
at the time the construction permit or operating permit applica-
tions are evaluated.
The importance of plant inspection as a field operations
activity that provides for the systematic detection and obser-
vation of emission sources cannot be overemphasized. The whole
process of inspection follows certain rules and guidelines which
are discussed briefly'in the following sections. Additional in-
formation regarding inspection procedures, field surveillance
and enforcement guidelines is presented in References 7 through 10.
5.1 CONDUCT OF INSPECTION
There are four important components in the conduct of in-
spection of a given equipment or a process.
• Formal Procedure (e.g. use of credentials, ask to see
appropriate official)
• Overall inspection process (e.g. review of process and
records)
• Safety precautions and procedures
• Frequency of inspection
5.1.1 Formal Procedure
Prior to actual on-site inspection, the inspector should
investigate any available data on plant operations, In prepara-
tion for the inspection the official should obtain the following
data:
* Information for each major source (from an air pollution
point of view) including process descriptions, flow
5-1
-------
diagrams, estimates of emissions, applicability of standards,
and previous related enforcement actions
• Plot plans showing disposition of all major units at the
faci1i ty
• Business and ownership data including names of responsible
management personnel
At the time of inspection, the inspector must have with him
the credentials showing his identity as an official of an air pollu-
tion control agency. He should arrange an interview with the
management of the refinery. The interview with refinery managers
and equipment operators can verify data gathered and clarify any
misunderstanding with regard to the information reviewed prior to
the inspection.
5.1.2 Overall Inspection Process
Some inspections, especially initial ones, are comprehensive,
designed to gather information on all equipment and processes of
the refinery. Others are conducted for specific purposes such as:
• Obtaining information relating to violations
• Gathering evidence relating to violations
• Checking permit or compliance plan status of equipment
• Investigating complaints
• Following up on a previous inspection
• Obtaining emissions information by source testing
An initial refinery inspection lays the groundwork for eval-
uating potential emissions of pollutants from a given source and
for assessing the relative magnitude of pollution control problems
requiring correction, reinspection, or further attention. The
initial inspection has two phases: a refinery survey and a
physical inspection of the equipment and processes. After this
inspection is complete, routine surveillance continues. Periodic
reinspections are scheduled and occasional special-purpose in-
spections (unscheduled) may be required.
During the initial survey, the inspector examines the pos-
sible effects of emissions on property, persons and vegetation
adjacent to the source; he may also collect samples or specimens
5-2
-------
that exhibit possible pollution-related damage. Sensory observations
(odor detection) are also made.
The NSP Standards for the facility were given in Section 2.0.
The details have been given in Section 2.2, where the records relat-
ed to the process which must be maintained were also discussed. The
inspector must also review these records kept by the operator.
An aid to the inspector is the information incorporated in
applications to operate the equipment. The permit status of the
equipment should be routinely checked to detect any changes in
equipment or process that might invalidate an existing permit or
conflict with variance conditions. Similarly, alteration of
equipment is frequently detected by discrepancies in the equipment
description or by changes noted on engineering applications in the
permit file.
5.1.3 Safety Equipment and Procedures
All refineries have standard safety procedures for employees
and visitors. These procedures also concern the inspector. The
inspector is accompanied to the unit or units to be inspected by
the air pollution representative within the plant or by such other
informed refinery personnel as he might indicate. Personal pro-
tection is necessary in many of the industrial locations that an
enforcement officer may be required to visit. The inspector
should wear a hard safety hat while in a plant. He should wear
rubber gloves and goggles when necessary. In the event of fire
in the area of inspection, the inspector must leave the plant
immediately, and remain outside the are^a until the "All Out" signal
is sounded. The inspector should be accompanied by another person
and two persons should remain together until the job is completed.
He must not smoke or carry cigarette lighters which ignite when
dropped within an oil refinery. He should use only approved
flashlights in oil refineries.
5.1.4 Frequency of Inspections
Because of the complexity of the petroleum industry, unit
processes must be inspected systematically and regularly, The
frequency of reinspection is based upon the findings during the
initial inspection and the recommendations of the inspector and his
supervisor. These recommendations obviously depend on whether or
not the "good" maintenance practices from the pollution standpoint
are followed by the operator. Further* the frequency would de-
pend on the overall inspection load of the control agency for the
whole district. The reinspections are scheduled so that they can
be completed within a month. The number of reinspections assigned
5-3
-------
per district is based on the estimate that all required inspections
can be completed within one year.
The enforcement officer may have occasion to inspect the
process out-of-schedule because of complaints or violations. In
these cases, he does not make a formal inventory reinspection,
but uses the copy of the previous inventory record (equipment list)
from his files as a check on status of the permit, compliance, or
other situation.
5.2 INSPECTION CHECKLIST
Data obtained during an inspection can be summarized on forms
similar to the ones shown on the ensuing pages. These forms also
serve as a record of inspection.
5.2.1 Facility Identification
Facility Name
Facility Address
Mailing Address
Telephone Number
Nature of Business
Date of Last Inspection.
Responsible Person to Contact
Persons Contacted at Plant
Site
Inspectors
Source Code Number
Date of Inspection
5-4
-------
5.2.2 Opacity Observations
Emission Source
Pre-entry observation:
Average percent equivalent opacity from Method 9 observation
Reading ranged from
to
Name of observer (inspector)
Post-entry observation:
% opacity.
5.2.3 Odors
Area
Area
Area
None Faint Strong (Circle one)
None Faint Strong (Circle one)
None Faint Strong (Circle one)
5.2.4 Acid Gas Treating Plant
a) Design input capacity (e.g. flow rate
of sour gas into the absorber)
b) Actual flow rate of sour gas into the
absorber
c) Output of unit (flow rate of treated
gas)
d) Actual flow rate of acid rich gas
to flare
cu m/min (cfm)
cu m/min (cfm)
cu m/min (cfm)
cu m/min (cfm)
5-5
-------
e) Actual flow rate of acid rich gas to
Sulfur recovery plant (or other unit)
cu m/min (cfm)
f) Concentration of
g)
in
Sour gas entering the
acid gas treatment plant
Treated gas going to
fuel gas system
Acid rich gas leaving
the' acid treatment plant
Disposal technique used Sulfur
for acid rich gases Plant
h) Any leakages?
(specify location)
i) Amine absorber's pressure
j) Type of absorber used
k) Concentration of absorber used
(e.g. 10 percent solution in
water)
1) Temperature of cool "lean"
amine entering the absorber
Flare
m) Temperature of hot "rich"
amine leaving the absorber
n) General Remarks
Condition of various temperature,
pressure, and flow ratio controls
and last maintenance date
mg/dscm (gr/dscf)
mg/dscm (gr/dscf)
mg/dscm (gr/dscf)
Other None
°C
General remarks on amine
stripper
Remarks on amine sump
Any isolated streams of sour gas
untreated for H2S removal and recovery?
(specify location)
Condition of filters (the cool "lean"
amine is filtered before entering
the absorber)
5-6
-------
5.2.5 Fuel Gas Combustion Unit
a) Type of unit (circle one)
b) Design capacity (size of
the unit)
c) Design fuel gas burn
rate
d) Actual fuel gas burn
rate
e) H2S concentration of
fuel gas
f) Heat content of fuel
gas
Boiler Process Heater
Flare Other (specify)
MM K-cal/hr (MM BTU/hr)
scm/hr (scf/hr)
scm/hr (scf/hr)
mg/dscm (gr/dscf)
Kcal/cu m (BTU/cf)
g) Is fuel gas treated in acid
gas treatment plant? Yes No
(Check one)
If answer to g) is No, only information for j) through r) must
be collected.
h) Location of H2S monitoring
equipment
i) Technique of HoS monitoring (e.g.
effective sampling)
j) Types and amount of auxiliary or additional fuels used (if any)
Natural Gas scm/hr (scf/hr)
liters/hr (gal/hr)
LT/hr (tons/hr)
specify units
Fuel Oils (specify number)
Coal
Other
k) Characteristics of auxiliary or additional fuels used (if any)
Fuel Used
Heat Content
Ash Content
Sulfur Content (Specify Units)
Natural Gas Kcal/cu m (BTU/cf)
Fuel Oil Kcal/liter (BTU/gal)
Coal Kcal/kg (BTU/lb)
Other (Specify Units)
mg
ing npo/ascm
(gr R2S/dscf)
_% S by wt.
_% S by wt.
_(Specify Units)
5-7
-------
1) Are exit gases from combustion unit treated for S02 removal?
Yes No (Check One)
m) Type of exit gas treatment (e.g.
wet scrubber) .
n) Location of SC^ monitoring
equipment
o) Quantity of S02 in
the exit gases kg/hr (Ib/hr)
p) Equivalent H2S concentration
in the fuel gases mg/dscm (gr/dscf)
q) Exit gas flow rate s cm/mi n (scfm)
acm/min (acfm)
r) Exit gas temperature °C (°F)
s) General remarks
(if any)
5.2.6 Air Pollution Control Equipment
Acid Gas Treatment Plant Wet Scrubber Other
A) Acid gas treatment plant (checklist already presented)
B) Scrubber
a) Evidence of corrosion and wear
b) Pressure drop across the tower
c) Type of scrubbing slurry used
(e.g. lime, limestone slurry)
d) L/G (liquid/gas) ratio
e) Effluent disposal system
f) Estimated collection
efficiency
g) Location of S02 monitoring
equipment
5-8
-------
h) Quantity of $03 in the
exit gases
t) Equivalent \-
fuel gases
concentration in
Kg/hr (Ib/hr)
mg/dscm (gr/dscf)
5.2.7 Records of Operation
a) Are records of operation kept? Yes
No
(Check one)
b) Quality of records (circle one) Good Fair Poor
c) Regulations regarding record keeping being followed? (see
Section 2.2.4 for regulations)
d) Initial Start-up
e) Startup, malfunction,
and shutdowns
f) Performance testing
g) Quarterly reports
5.2.8 Operational Aspects
a) Plant operating within specified
limits?
Yes No If No, describe
b) Any changes or modifications to
equipment? (modifications without
Agency's approval?) Yes
c) Evidence of lack of maintenance Yes
d) Location of continuous monitoring
equipment for measuring H2S
concentration in the treated
fuel gas
e) Other general remarks
(specify)
No If Yes, describe
No If Yes, describe
5-9
-------
5.3 INSPECTION FOLLOW-UP PROCEDURES
After the completion of the inspection, the inspector must
determine the compliance status of the source based on the data
collected. If the collected data are not sufficient for the
compliance evaluation, he must contact appropriate refinery
personnel to acquire required data. As a precaution, the validity
of data must be checked. If an inspection indicates that a source
is not operating in compliance with applicable regulations, the
inspector should follow established agency procedures regarding
notice of violation, request for source test, and related matters.
The various items which could result in a determination of
non-compliance would include:
• Concentration of HgS in excess of 230 mg/dscm (0.1 gr/dscf)
in the fuel gas after acid gas treatment which is burned
in the fuel gas combustion device without any after treat-
ment for S02 removal.
• Concentration of S02 in excess of equivalent H^S concen-
tration of 230 mg/dscm (0.1 gr/dscf) in the fuel gases
resulting from the combustion of untreated fuel gas and
after appropriate control method.
• Monitoring equipment for H2S in fuel gas or for S02 in
exit gases not in operation.
The inspector should check to ensure that permits have been
granted for all applicable processes and equipment and their modifi-
cations. For any later public complaints, he should determine
cause of complaint, record pertinent data, issue violation notices
if appropriate, and ascertain adequacy of plans for prevention
of future incidents. He periodically should review emergency
procedure plans and make sure that all shut-down procedures are
being implemented during periods of process curtailment while
coordinating with other agencies participating in pollution
reduction efforts. As a part of inspection follow-up procedures,
he should also check to see that engineering, procurement, installa-
tion, and testing of equipment is proceeding according to the
approved plan.
In the case of incident and complaint investigations, court
actions, and variance board activity, the inspector will need the
data collected during his previous inspection visits of the facility.
5-10
-------
For example, the point of emission of excessive odors may be traced
from an incident described in an operator's log or from an odor
survey record.
5-11
-------
6.0 PERFORMANCE TEST
The NSP Standards require a performance test of any new fuel
gas combustion device. In order to guarantee the validity of the
test, an inspection team will be present at the facility for obser-
vation. The team should consist of three enforcement personnel
with the following areas of responsibility during test period.
• Monitor process operating conditions from the control room.
• Make visible observations of process operations from the
plant area.
• Monitor emission testing procedures from the test site.
Each team member will fill out checklist type data during the test
and will submit a report including analysis of the data and indication
of any upset conditions which may have affected the test.
6.1 PROCESS OPERATING CONDITIONS
For the purpose of obtaining source test data which is truly
representative of the operating characteristics of the Acid Gas
Treatment Plant or the fuel combustion device being tested, it is
extremely important that the test be conducted at or above the
maximum production rate at which the unit will normally be operated.
In certain cases, the EPA may feel that conditions other than the
maximum operating rate of the plant should be used to achieve valid
test results. In such cases, the EPA will designate conditions at
which source performance testing is to take place. In all cases,
inspectors must personally verify that the plant is operating at
the specified conditions, and that a stabilized, steady state of
operation has been reached. Such verification should be made with
the plant operator and refinery manager, and inspectors should
observe process controls (i.e. gauges, rate-meters, and recorders)
to determine that operating conditions are as designated. During
the source test, inspectors should periodically check operating
conditions of the plant, carefully noting any changes in the operating
parameters such as temperature, pressure, or sour gas flow rate.
Since the NSP standards apply to the fuel gas which is to be
burned in a combustion device, inspectors must make certain that
the sampling is done at the point where the treated gas is provided
to the combustion unit or, if all gases are treated, at the point
where the gas is released to the refinery fuel supply line. As
most refinery fuel gas lines are operated at rather high pressures,
it is important to ascertain that proper precautions are used in
sampling these lines, including the use of pressure reduction valves
where necessary.
6-1
-------
In the event that the owner or operator Is using untreated
fuel gas in the combustion device and is treating the exit gases
from the combustion device in an appropriate manner to control
the S02 emissions within the specified limits of the regulation,
inspectors should make certain that the sampling is done for the
S02 emissions at the point where exit gases are discharged from
the air pollution control system or at the point of discharge
from the combustion device if no control system is present.
Inspectors must verify that provisions exist for logging and
recording the operating rate as well as the hours of operation of
the units concerned on a daily basis.
6.2 PROCESS OBSERVATIONS
6.2.1 Acid Gas Treatment Plant
Untreated refinery fuel gas contains appreciable quantities
of hydrogen sulfide (H£S) and must therefore be treated to remove
this component prior to the burning of the refinery gas in process
units, in order to prevent formation of excessive emissions of
sulfur dioxide. Acid Gas Treatment Plants remove C02 and H2S from
the refinery fuel gas stream by absorption in an absorbing solution.
The solution is then separated from the de-sulfurized refinery fuel
gas and the H2S and C02 are reclaimed by heating the absorbing
solution under reduced pressure. The H2$ and C02 components may
then be sent to a sulfur recovery plant for reclamation, while the
regnerated absorbing solution is recycled.
Since the Acid Gas Treatment Plant is a closed system, no
pollution control devices are installed as a general rule, however,
inspectors should make sure that samples are taken from the centroid
(or a point near the center) of the fuel gas line at a point just
after the treated gas is released from the plant.
At the time the performance test is conducted, the inspector
should make careful notation as to the existing layout of the treat-
ment plant, type of absorbing solution, and operating parameter ranges
of the plant (such as pressures, temperatures and flow rates).
Inspectors should also determine that a provision exists for monitor-
ing (by instrumental means) the concentration of hydrogen sulfide
in the treated fuel gas just after it is released into the refinery
fuel gas line. If possible, a photographic record should be made
of the principle process units as any modifications to the plant
will likely change the results of future performance tests. Inspec-
tors should determine what means of disposal are utilized for the
H;?S which has been separated in the treatment plant (most often it
will be transported to a sulfur recovery plant for removal of
elemental sulfur). The equipment checklist is the same as that which
appears in Section 5.2.
6-2
-------
6.2.2 Fuel Gas Combustion Device
The inspector should first check the overall configuration
of the combustion devices (i.e. to check if they have acid gas
treatment plant or if they have a pollution control system for
the exit gases). In any event, he should note the type of any
other additional or auxiliary fuels that the refinery may be
burning along with the fuel gas. General information regarding
the combustion unit such as type of equipment and other parameters
as presented in Section 5.2 of this report should be carefully
observed.
Inspectors should also determine that a provision exists
for monitoring (by instrumental means) the H2S concentration in
the treated fuel gas just after it is released into the refinery
fuel gas line. Inspectors should also check any S02 monitoring
equipment downstream of the combustion unit or pollution control
system, whichever may be the case, when untreated gases are burned
in the combustion device. As a general observation practice for
combustion units, inspectors should check and observe the stack
associated with the combustion device for possible visible
contaminants. Often the presence of a detached, visible plume
indicates a sulfuric acid or SOg mist emission. Operating
parameters and control of these parameters for the pollution
control system should be carefully checked.
6.3 EMISSION TEST OBSERVATIONS
Emission source testing discussed here concerns determining
compliance of new sources with EPA New Source Performance Standards
(NSPS). During source testing operations, field inspectors should
periodically spot check testing procedures, equipment, and data to
make certain that the test is valid.
All performance tests should be conducted while the unit being
tested is operating at or above maximum production rate at which
unit will normally be operated. If EPA Administrator feels that
other conditions should be used to achieve valid test results,
such will be used as basis for testing.
6.3.1 Traversing (EPA Method No. 1, Reference 6)
Of first importance is the selection of a sampling point and
determination of a minimum number of traverse points to ensure the
collection of a representative sample. Inspectors should make
certain that the sampling site selected is a minimum of eight (8)
diameters downstream and at least two (2) diameters upstream from
any disturbance to the flow of gases within the duct or stack which
6-3
-------
is being sampled. Such disturbances are commonly caused by
expansions or contractions, bends, visible flames, observable
cross-members, or other entering ducts.
6.3.2 Stack or Duct Gas Velocity Determination (EPA Method No. 2,
Reference 6)
In the determination of gas velocity within the duct or
stack, it should be made certain that all data from each traverse
point is carefully and accurately recorded as this is the basic
information used to determine the stack flow rate. Each point deter-
mined by Method 1 shall be identified by a number and the following
information shall be recorded: Velocity head in inches of water;
stack (duct) pressure in inches of Mercury; and temperature (unless
the total temperature variation with time is less than 10°C. Care
should be taken to determine that a type "S" pi tot tube is used to
obtain the velocity head readings and that this tube is of sufficient
length to reach all traverse points. The pi tot tube should be
graduated with temporary markings (i.e., tape or chalk marks) such
that each traverse point may be reached by successively moving the
tube deeper into or withdrawing it further from the duct or stack
being sampled. All tubing and connectors between the pitot tube
and the inclined manometer or draft gauge should be tight and leak-
free. An inclined manometer or draft gauge should be used to obtain
velocity head readings from the pitot tube. Make certain that this
gauge is filled with sufficient colored liquid to give readings
throughout its range of calibration, and that the manometer liquid
level is adjusted to read "zero" with the end of the pitot tube
shielded from incidental breezes prior to beginning the velocity
head measurement. Periodically check to make sure that no constric-
tion occurs in the hose connections during the course of the velocity
head measurement.
The most common means of stack temperature measurements is by
thermocouple and potentiometer; operation of this equipment is rather
straight forward although several points should be checked to ensure
accurate measurement. The thermocouple connecting wires should be
securely tightened to the terminal lugs on the potentiometer, and
it should be determined that the thermocouple circuit is complete
(an open circuit will be evident if the potentiometer fails to
balance, giving readings off the scale of the instrument). If the
potentiometer being used is not an automatic compensating type
(automatic reference to ambient temperature), see that the ambient
air temperature has been recorded or that the potentiometer scale
has been calibrated with this temperature as a reference. While
taking gas temperature readings, sufficient time should be allowed
(normally about five minutes) for the thermocouple probe to reach
thermal equilibrium with the duct gas before taking the first few
readings.
6-4
-------
As part of the data necessary for the velocity determination,
the static pressure within the stack or duct should be measured.
This is done using a mercury-filled "U" tube manometer, one end of
which is open to the atmosphere and the other connected to a probe
extending into the duct or stack itself. Again, the tubing from
the probe to the manometer must be free of constrictions and
tightly connected at both ends. A barometric pressure reading
(of atmospheric pressure) in inches of mercury, should be obtained
from a standard barometer located in the general vicinity of the
test site; this can be a wall-mounted barometer in the plant offices,
laboratory, or any convenient location which is at ambient tempera-
ture and free to vibration.
6.3.3.Hydrogen Sulfide (EPA Method No. 11. Reference 2 and 6)
Hydrogen sulfide (H2S) is sampled from the centroid (or a
point near the center) of the fuel gas line. Inspectors should
check to see that the probe is in the required position and that
a sampling valve is placed in the probe just after it emerges from
the refinery fuel gas line. A word of caution is in order here
as the fuel gas line is often operated at high pressure and a
pressure reducing valve probably will be found where the sample
is to be taken, care should be exercised in opening this valve.
The sample line immediately after the valve must be 1/4 inch
Teflon tubing with provisions for heating in order to eliminate
condensation of moisture from the sample as it is extracted from
the gas line. The Teflon tubing must connect to a series of
five (5) midget impingers (30 ml capacity each) which are fitted
with ball and socket joints. Several points are important to
check in connection with the impinger series: all connections
must be air-tight and the ball joint connections must be greased
and secured with proper size retaining spring-clips; 15 ml of 3
percent hydrogen peroxide should be placed in the first impinger;
absorber solution must be placed in the next three (3) impingers
(15 ml of the solution in each) with the fifth remaining empty;
an ice water bath must be in contact with the impingers in such
a manner as to maintain the last (dry) impinger at 20°C (70°F).
The absorbing solution has a tendency to settle out; be sure that
this solution in the impingers is well in-suspension before
sampling begins.
6-5
-------
Immediately after the impinger series is a silica gel drying
tube followed by a needle valve, pump (pump can be eliminated if
refinery fuel gas line has sufficient pressure to drive gas through
the sampling train), rotameter, and finally a dry gas meter. The
sample line should be purged and the initial sampling rate should be
set at 1.13 liters per minute.
No samples shall be taken in less than 10 minutes and some
fuel gas lines in which the concentration of H2S is low will require
longer sampling periods. Bear in mind that the collection of H2S
is evident by formation of the yellow CdS precipitate. Minimum
sample volume is 0.01 dry standard cubic meters (0.35 dscf) per
sample. The arithmetic average of two determinations constitutes
one run, each determination being made at approximately one-hour
intervals.
The following data should be recorded by inspectors in order
to check results later: sampling time, gas volume collector; rota-
meter setting; and meter setting for each sample.
Before sampling, the line between the probe and the first
impinger should be purged. Observe sample recovery methods and
cleaning procedures between samplings. Make sure that the acidified
iodine solution is used to wash out impingers and connecting glass-
ware and that the washings are transferred directly into the iodine-
number flask. The flask itself must be kept tightly stoppered at
all times except when adding washings and this must be completed
as rapidly as possible to avoid loss of iodine vapor.
6.3.4 Sulfur Dioxide (EPA Method No. 6, Reference 6)
This method is applicable for the determination of sulfur
dioxide emissions from stationary sources when specified by the
test procedures for determining compliance with New Source
Performance Standards. Method 1, as discussed in Section 631,
is used for velocity traverses and Method 2 (Section 632) for
determining velocity and volumetric flow rate. The sampling site
for determining S02 concentration by Method 6 is the same as for
determining volumetric flow rate by Method 2.
6-6
-------
The sampling point in the duct for determining S02 concen-
tration by this Method 6 is at the centroid of the cross section
if the cross sectional area is less than 5 m2 (54 ft2) or at a
point no closer to the walls than 1 m (39 inches) if the cross
sectional area is 5 m2 or more and the centroid is more than one
meter from the wall. Inspectors should check to see that the
probe is in the required position. Inspectors should also check
to see if the Pyrex glass probe is equipped with heating system
to prevent condensation and a filtering medium to remove particulate
matter including sulfuric acid mist.
The sample line is connected to a midget bubbler and three
midget impingers in series. The midget bubbler should be checked
for the glass wool packing in the top part to prevent sulfuric
acid mist carryover. Several points are important to check in
connection with the impinger series: all connections must be air-
tight and the ball joint connections must be secured with proper
size retaining spring-clips; 15 ml of 80% isopropanol must be
placed into the midget bubbler and 15 ml of 3% hydrogen peroxide
into each of the first two midget impingers with the fourth one
dry (remaining empty); an ice water bath must be in contact with
the impingers in such a manner so as to keep the temperature of
the gases leaving the last impinger at 21°C (70°F) or less.
Inspectors should make sure that a leak check test run of
the sampling train at the sampling site is conducted by plugging
the probe inlet and pulling a 10 inches Hg vacuum. After plugging
or pinching off the outlet of the flowmeter, the pump should be
turned off. The vacuum should remain constant for one minute.
Immediately after the impinger series is a silica gel drying
tube followed by a pump, needle valve, rotameter, and finally a
dry gas meter along with a thermometer. The sample is extracted
at a rate proportional to the gas velocity at the sampling point.
No samples shall be taken in less than 10 minutes and the minimum
sampling volume shall be 0.01 dscm (0.35 dscf) for each sample.
The arithmetic average ofJtwo samples shall constitute one run.
Samples shall be taken at approximately 1-hour intervals.
6-7-
-------
The following data should be recorded by the inspectors in
order to check the results later:
• Sample volume measured by the dry-gas meter, cu m (cu ft)
• Average dry gas meter temperature, °K (°R)
• Barometric pressure at the orifice meter, mm Hg (inch Hg)
The inspector should also observe the sample recovery methods and
cleaning procedures between samplings. The contents of the midget
bubbler is discarded while the contents of the midget impingers
are poured into a polyethylene shipment bottle. The midget_
impingers and the connecting tubes are rinsed with the distilled
water and these washings are added to the same storage container.
6.4 EMISSIONS MONITORING
6.4.1 H2S Monitoring
No specific method for HnS emission monitoring has been speci-
fied by EPA, however, continuous monitoring for H2S in acid gas treat-
ment plants can be accomplished by a monitor in the treated gas line
(return of treated gas to refinery fuel gas supply). Since hydrogen sul-
fide monitoring is a new technique, the majority of Acid Gas Treatment
Plants will not be equipped with continuous monitoring instrumentation
for hydrogen sulfide.
On those installations which are so equipped, the technique
of extractive sampling will be used. Of those systems which are
in use, the electrochemical cell is most widely used, followed by
flame photometric (specific for sulfur). Currently no practical
in-situ methods are employed.
The sample extraction portions of all of these systems will
be basically alike and, all should be equipped with filtering
devices to eliminate particulate matter and moisture. The probe
should be constructed of Teflon, Vycor, or stainless steel, but
not of reactive materials such as copper, brass, or aluminum.
Provision should be made for periodic purging and/or cleaning
of the sampling probe. Any optical surfaces or mirrors must be
kept clean and in good repair. A regular routine maintenance
program must be carried out in order to keep the system operating
at its designed level of reliability.
The analytical or detector portions of the various systems
are slightly different from one another in that the electrochemical
cell is a small, "black box" unit which is essentially self-contained,
6-8
-------
while the flame photometric type is a larger unit which depends
upon external fuel supplies in order to carry out its analysis.
The electrochemical cell utilizes a semi-permeable membrane
through which a portion of the H2S passes into the cell. In
the cell, electro-oxidation of the gas produces an electric signal
at a rate directly proportional to the concentration of the gas
in the sample.
The flame photometric system uses a photomultiplier tube and
filter specific for detection of excited short-term sulfur emissions.
The sample itself is burned to achieve the necessary excitation for
the detector tube. This method responds positively to other sulfur-
containing compounds. The detector should be checked for cleanliness
of the optical surfaces as well as the adequacy of supply and quality
of the fuel gas. The burner in this type of detector is especially
prone to gathering dirt and ash, so extra care should be taken to
see that these areas are regularly cleaned. Inspection of electro-
chemical cell detectors should include checking the completeness of
the cell (particularly the semi-permeable membrane and gasketing).
Data recording devices should be equipped with an alarm
provision to sound when H2S level reaches a predetermined level,
in this case 230 mg/dNm3 (0.10 gr/dscf). The detector portions
of these monitoring systems should be located fairly close to
the sampling site, but should be free of vibration, heat, shock,
protected from the weather, and should provide safe convenient
access for purposes of calibration and maintenance.
Calibration must be carried out at least once in every 24
hour period unless the manufacturer suggests a more frequent
calibration interval, in which case his suggestions must be
followed. If possible, inspectors should arrange for a demon-
stration of the alarm system by "doping" the sampling inlet with
a heavy concentration of H2S, for the purpose of determining
that the feature is effective.
If a monitoring system is encountered which is not covered
in this manual, inspectors should first ascertain the principle
involved in its operation. With this information at hand, an
effective inspection can then be made of the sampling and analysis
6-9
-------
portions of the system. The data recording and alarm systems will
be basically alike. A few points to check are listed as follows:
1) Probe material; make certain it is a non-reactive
material and that adequate provision exists for
cleaning and/or purging the probe.
2) Any optical surfaces should be in some manner
protected from direct exposure to stack gases and,
a filtering system to exclude moisture and particulate
material should be included in the sample extraction
portion.
3) The analytical portion of the system should be so lo-
cated as to minimize its exposure to vibration, shock
weather, and dirt. Safe, convenient access should be
provided for purposes of calibration and maintenance.
4) Inspectors should determine frequency of calibration,
cleaning, and routine maintenance.
5) Data recording system must be operational and
an alarm feature should be incorporated to sound
when pollutant being monitored reaches a
predetermined level. A demonstration of this
feature in operation should be observed.
6.4.2 SO? Monitoring
The direction of current developments indicates that common
usage for S02 emission monitoring may settle on certain types of
automatic physical-chemical and electro-optical instrumentation.
There are three basic approaches to source monitoring that could
foreseeably meet the current requirements of the EPA.
• Extract a continuous sample from the stack or duct
and feed it to any appropriate analytical instrument,
after initial conditioning by means of a sample
handling/interface system.
• Observe and analyze the stack gas in situ by means of
an optical instrument whose light path traverses the
gases inside the stack or duct.
• Observe and analyze the stack gases shortly after they
leave the stack, by means of a remote optical instrument.
6-10
-------
In the extractive approach, a continuous sample is drawn
from the stack and transported to the analyzer, which can be
mounted in any convenient location. This requires a probe mounted
in the stack or duct, and some form of interface system to provide
the analyzer with a sample that is in an appropriate state of
cleanliness, temperature, pressure, and moisture content. This
approach is the oldest and has provided the most experience to
date.
With the in situ approach, the instrument is mounted either
inside, or just outside the stack. In the case of photometric
and spectroscopic instruments, the light source may be mounted
on one side of the stack and the detector on the other, or else
the instrument may incorporate an extended mechanical beam with
a mirror on the end or a mirror may be mounted on the opposite
side of the stack so that the light penetrates a fixed distance
and is returned to the instrument on the same side. Since these
spectroscopic instruments determine pollutant concentrations by
their discrete spectral absorption, they must be capable of
discriminating against absorption and scattering of the light
by particulates. They must also have high discrimination against
all other, unwanted, components that are present in the stack gas,
such as water vapor or carbon dioxide.
The in situ approach provides an average reading across
the whole stack which may be considered an advantage over the
point sampling approach that is most common with the extractive
systems. It is possible, however, to provide the averaging
function with the extractive system by using multiple integrat-
ing probes.
One problem facing all in situ optical instruments is that
of keeping the optical windows clean. The common solution is to
bathe the windows with a stream of clean air. This appears to
be fairly effective for a reasonable period of time, but periodic
mechanical cleaning is required. Another problem with the in
situ instruments is to provide a satisfactory method of checking
zero and span periodically- This is normally done with most ana-
lytical instruments by supplying known blends of zero and span
gases to the instrument. This is not as practical with in situ
monitors, as it is with extractive systems.
In order to check the S02 monitoring instrument, inspectors
are recommended to collect information prior of the inspection
regarding the type of S02 monitoring instrument used including a
manual about the instrument-system and study the problems
associated with the system used. As a general practice, to
ensure that all of the apparatus is working properly, stack
6-IT
-------
gas results obtained from EPA Method 6 should be compared to those
from monitoring data. Instruments and sampling systems installed
and used shall meet specifications prescribed by the Administrator
and each instrument shall be calibrated in accordance with the
method prescribed by the manufacturer of the instrument. Inspectors
should check to see if zero adjustment is done according to manu-
facturer's recommended procedure and that calibration procedures
are performed at least once per 24-hour operating period or at
short intervals as specified by the manufacturer.
6-12
-------
LIST OF REFERENCES
1. "State Implementation Plan Emission Regulations for Sulfur
Oxides: Fuel Combustion.", EPA-450/2-76-002, March 1976.
2. Federal Register, Volume 39, Number 47, March 8, 1974.
3. Federal Register, Volume 38, No. Ill, June 11, 1973.
4. Kohl, A.L., F.C. Riesenfeld, "Gas Purification", McGraw-
Hill, New York, 1960.
5. Stecher, P.G., "Hydrogen Sulfide Removal Processes",
Noyes Data Corporation, Park Ridge, New Jersey, 1972.
6. Federal Register, Volume 36, Number 247, December 23, 1971.
7. Weisburd, M.I., "Air Pollution Control Field Operations Manual,
A Guide for Inspecting and Enforcement", Department of Health,
Education and Welfare, Public Health Service, Division of Air
Pollution, Washington, D.C., Publication Number 937, 1962.
8. Brandt, C.S., and W.W. Heck, "Effects of Air Pollutants on
Vegetation", in "Air Pollution", Vol. 1, Stern, A.C. (ed.),
Academic Press, New York, 1968.
9. "Guide for Compiling a Comprehensive Emission Control Inventory
(revised)", Environmental Protection Agency, Research Triangle
Park, N.C., Publication Number APTD-1135, March, 1973.
10. "Field Surveillance and Enforcement Guide for Petroleum
Refineries (Final Draft)", prepared by the Ben Holt Co.,
Pasadena, California for Environmental Protection Agency,
Research Triangle Park, N.C., July, 1973.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
. REPORT NO.
EPA 340/1-77-019
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Inspection Manual for the Enforcement of New Source
Performance Standards: Fuel Gas Fired Combustion Units
5. REPORT DATE
August. 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
George E. Umlauf, Bansi Parekh
8. PERFORMING ORGANIZATION REPOI
9. PERFORMING ORGANIZATION NAME AND ADDRESS
PACIFIC ENVIRONMENTAL SERVICES
1930 14th Street
Santa Monica, CA. 90404
10. PROGRAM ELEMENT NO.
Task Order #26
11. CONTRACT/GRANT NO.
68-01-3156
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
Division of Stationary Source Enforcement
401 "M" Street, S.W.
Washington, D.C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This document outlines the air pollution requirements of Federal New Source Perfor-
mance Standards for petroleum refinery combustion units which are fired on fuel gas.
It is designed to be utilized by agency personnel who are responsible for the enforce-
ment of these regulations. Briefly, the rules are applicable to all combustion units
on which construction or modification was commenced after June 11, 1973. If fuel gas
is burned in the unit, it is required that the fuel gas be desulfurized to a level
of 230 mg/dscm (0.1 gr/dscf) or below, or that effluent combustion gases be treated
for sulfur dioxide removal to an equivalent level. This manual also outlines other
requirements of the regulation for performance testing, monitoring and record keeping
and reporting requirements. Subsequent sections of the document describe types of
emission control systems which may be used; procedures for conducting on-site inspec-
tions including an inspector's checklist; and approved methods for conducting
performance tests.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
Air Pollution
Refinery
Combustion
Desulfurization
Sulfur Dioxide
New Source Performance
Standards
Fuel Gas
Emission Control Systems
Inspections
0701
1302
1308
2102
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
47
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
ft U.S.GOVERNMENT PRINTING OFFICE: 1977- 260-880:83
------- |