1. Contents:
    2. Tables:
    3. Figures:
    4. Abbreviations and Acronyms:
    5. Executive Summary:
    6. Introduction:
    7. Definitions:
    8. Applicability/Scope:
    9. Prohibited Use of GP-5 and GP-5A:
    10. Authorization to Use GP-5 or GP-5A:
    11. General Permit Fees:
    12. Applicable Laws:
      1. Natural Gas Production, Transmission, and Distribution for which
    13. Compliance Requirements and Compliance Certification:
    14. Notification Requirements:
    15. Recordkeeping Requirements:
    16. Reporting Requirements:
    17. Source Testing Requirements:
    18. General Methodology of Determining Best Available Technology:
      1. Oxides of Nitrogen
      2. Carbon Monoxide
      3. Volatile Organic Compounds
      4. Hazardous Air Pollutants
      5. Oxides of Sulfur
      6. Particulate Matter
      7. Carbon Dioxide
      8. Methane
    19. Sources Common to GP-5 and GP-5A:
      1. Fugitive Particulate Matter
      2. Natural Gas-Fired Combustion Units
      3. Stationary Natural Gas-Fired Spark Ignition Internal Combustion Engines
      4. Emissions from Lean-Burn and Rich-Burn Engines
      5. Emission Control Technology
      6. Combustion Control
      7. Post-Combustion Emission Reduction Technology for Rich-Burn Engines
      8. Post-Combustion Emission Reduction Technology for Lean-Burn Engines
      9. Engine Size Grouping
      10. Engine Emission Limits
      11. Rich-Burn Engines Less Than 100 bhp
      12. Rich-Burn Engines Greater Than or Equal To 100 bhp but Less Than 500 bhp
      13. Rich-Burn Engines Greater Than or Equal To 500 bhp
      14. Lean-Burn Engines Less Than 100 bhp
      15. Lean-Burn Engines Greater Than or Equal To 100 bhp but Less Than 500 bhp
      16. Lean-Burn Engines Greater Than or Equal To 500 bhp but Less Than 1,875 bhp
      17. Lean-Burn Engines Greater Than or Equal To 1,875 bhp
      18. Reciprocating Natural Gas Compressors
      19. Existing Reciprocating Natural Gas Compressors
      20. New Reciprocating Natural Gas Compressors
      21. Glycol Dehydration Units and Associated Equipment
      22. Emission Limits for Glycol Dehydrators
      23. Existing Glycol Dehydrators
      24. New Glycol Dehydrators
      25. Storage Vessels
      26. Emission Limits for Storage Vessels
      27. Existing Storage Vessels
      28. New Storage Vessels
      29. Tanker Truck Load-Out Operations
      30. Fugitive Emissions Components
      31. Controllers
      32. Pumps
      33. Enclosed Flares and Other Control Devices
      34. Pigging Operations
    20. Sources Specific to GP-5A:
      1. Well Drilling and Hydraulic Fracturing Operations
      2. Well Completion Operations
      3. Wellbore Liquid Unloading
    21. Sources Specific to GP-5:
      1. Natural Gas-Fired Simple Cycle Turbines
      2. Emissions from Natural Gas-Fired Turbines
      3. Emission Control Technology
      4. Combustion Control
      5. Post-Combustion Emission Reduction Technology for Turbines
      6. Turbine Size Grouping
      7. Turbine Emission Limits
      8. Turbines Rated Greater Than or Equal To 1,000 bhp but Less Than 5,000 bhp
      9. Turbines Rated Greater Than or Equal To 5,000 bhp but Less Than 15,900 bhp
      10. Turbines Rated Greater Than or Equal To 15,900 bhp
      11. Centrifugal Natural Gas Compressors
      12. Existing Centrifugal Natural Gas Compressors
      13. New Centrifugal Natural Gas Compressors
      14. Natural Gas Fractionation Process Units
      15. Sweetening Units
    22. Appendix A – Average Gas Composition Analysis
    23. Appendix B – SCR Cost Analysis for Engines and Turbines
    24. Appendix C – Oxidation Catalyst and NSCR Cost Analysis for Engines and Turbines
    25. Appendix D – Cost Analysis for Combustion Control Devices
    26. Appendix E – LDAR Cost Analysis
      1. The Department’s Independent LDAR Cost Analysis
    27. Appendix F – Urbanized Areas and Urban Clusters from the 2010 Census
    28. Appendix G – Non-road Engine Standards
      1. General Emission Standards Severe-Duty Engine Emission Standards
      2. NOX + HC CO NOX + HC CO

02/04/2017
Page
1
of
73
Bureau of Air Quality
Technical Support Document
General Plan Approval and General
Operating Permit for Unconventional
Natural Gas Well Site Operations and
Remote Pigging Stations (BAQ-
GPA/GP-5A) and for Natural Gas
Compressor Stations, Processing
Plants, and Transmission Stations
(BAQ-GPA/GP-5)
February 4, 2017

02/04/2017
Page
2
of
73
Contents:
Abbreviations................................................................................................................................................ 5
Executive Summary ...................................................................................................................................... 7
Introduction................................................................................................................................................... 7
Definitions..................................................................................................................................................... 8
Applicability/Scope..................................................................................................................................... 10
Prohibited Use of GP-5 and GP-5A............................................................................................................ 11
Authorization to Use GP-5 or GP-5A ......................................................................................................... 12
General Permit Fees.................................................................................................................................... 12
Applicable Laws ......................................................................................................................................... 12
Compliance Requirements and Compliance Certification.......................................................................... 14
Notification Requirements .......................................................................................................................... 14
Recordkeeping Requirements ..................................................................................................................... 15
Reporting Requirements ............................................................................................................................. 15
Source Testing Requirements ..................................................................................................................... 15
General Methodology of Determining Best Available Technology ........................................................... 16
Oxides of Nitrogen.................................................................................................................................. 17
Carbon Monoxide ................................................................................................................................... 17
Volatile Organic Compounds ................................................................................................................. 18
Hazardous Air Pollutants ........................................................................................................................ 18
Oxides of Sulfur...................................................................................................................................... 18
Particulate Matter.................................................................................................................................... 19
Carbon Dioxide....................................................................................................................................... 19
Methane .................................................................................................................................................. 20
Sources Common to GP-5A and GP-5 ....................................................................................................... 21
Fugitive Particulate Matter ..................................................................................................................... 21
Natural Gas-Fired Combustion Units ..................................................................................................... 21
Stationary Natural Gas-Fired Spark Ignition Internal Combustion Engines........................................... 23
Emissions from Lean-Burn and Rich-Burn Engines........................................................................... 23
Emission Control Technology ............................................................................................................ 23
Engine Size Grouping ......................................................................................................................... 25
Engine Emission Limits...................................................................................................................... 25
Reciprocating Natural Gas Compressors ................................................................................................ 32
Existing Reciprocating Natural Gas Compressors.............................................................................. 33
New Reciprocating Natural Gas Compressors.................................................................................... 33
Glycol Dehydration Units and Associated Equipment ........................................................................... 33
Emission Limits for Glycol Dehydrators ............................................................................................ 34
Storage Vessels ....................................................................................................................................... 35
Emission Limits for Storage Vessels .................................................................................................. 36
Tanker Truck Load-Out Operations........................................................................................................ 36
Fugitive Emissions Components............................................................................................................. 37
Controllers............................................................................................................................................... 39
Pumps...................................................................................................................................................... 39
Enclosed Flares and Other Control Devices ........................................................................................... 40
Pigging Operations.................................................................................................................................. 41
Sources Specific to GP-5A ......................................................................................................................... 42

02/04/2017
Page
3
of
73
Well Drilling and Hydraulic Fracturing Operations ............................................................................... 42
Well Completion Operations .................................................................................................................. 43
Wellbore Liquid Unloading .................................................................................................................... 43
Sources Specific to GP-5 ............................................................................................................................ 44
Natural Gas-Fired Simple Cycle Turbines.............................................................................................. 44
Emissions from Natural Gas-Fired Turbines ...................................................................................... 44
Emission Control Technology ............................................................................................................ 45
Turbine Size Grouping........................................................................................................................ 46
Turbine Emission Limits..................................................................................................................... 46
Centrifugal Natural Gas Compressors .................................................................................................... 49
Existing Centrifugal Natural Gas Compressors .................................................................................. 50
New Centrifugal Natural Gas Compressors........................................................................................ 50
Natural Gas Fractionation Process Units ................................................................................................ 50
Sweetening Units .................................................................................................................................... 51
Appendix A – Average Gas Composition Analysis.................................................................................... 53
Appendix B – SCR Cost Analysis for Engines and Turbines..................................................................... 54
Appendix C – Oxidation Catalyst and NSCR Cost Analysis for Engines and Turbines ............................ 59
Appendix D – Cost Analysis for Combustion Control Devices ................................................................. 65
Appendix E – LDAR Cost Analysis ........................................................................................................... 66
Appendix F – Urbanized Areas and Urban Clusters from the 2010 Census............................................... 71
Appendix G – Non-road Engine Standards................................................................................................. 73
Tables:
Table 1 - Engine Source Testing Requirements ........................................................................... 16
Table 2 - BAT Emission Limits for Natural Gas-Fired Combustion Units.................................. 22
Table 3 - BAT Emission Limits for Existing SI-RICE................................................................. 30
Table 4 - Proposed BAT Emission Limits for New SI-Rice ........................................................ 30
Table 5 - 40 CFR Part 60 Subpart JJJJ Requirements .................................................................. 31
Table 6 - 40 CFR Part 63 Subpart ZZZZ Requirements............................................................... 32
Table 7 - Glycol Dehydrator Control Thresholds and Control Requirements.............................. 35
Table 8 - BAT Emission Limits for Existing Turbines................................................................. 48
Table 9 - Proposed BAT Emission Limits for New Turbines ...................................................... 49
Table 10 - Methane de Minimus Calculations.............................................................................. 53
Table 11 - SCR Cost Analysis for 1,380 HP Engine .................................................................... 54
Table 12 - SCR Cost Analysis for 4,735 hp Engine ..................................................................... 55
Table 13 - SCR Cost Analysis for Turbines ................................................................................. 56
Table 14 - Calculated Cost per Ton of NO
X
Reduced vs Average Cost per Ton NO
X
Reduced for
Engines.............................................................................................................................. 57
Table 15 - Calculated Cost per Ton NO
X
Reduced vs Average Cost per Ton NO
X
Reduced for
Turbines using BAT Emissions ........................................................................................ 58
Table 16 - Oxidation Catalyst Cost Analysis for Lean-Burn Engines - Weighted Average
Emissions .......................................................................................................................... 59
Table 17 - Oxidation Cost Analysis for Lean-Burn Engines - BAT Emissions ........................... 59
Table 18 - NSCR Cost Analysis for Rich-Burn Engines - Weighted Average Emissions ........... 60

02/04/2017
Page
4
of
73
Table 19 - NSCR Cost Analysis for Rich-Burn Engines - BAT Emissions ................................. 60
Table 20 - Turbine Characteristics and Emissions Data............................................................... 61
Table 21 - Oxidation Catalyst Cost Data for Turbines ................................................................. 62
Table 22 – Oxidation Catalyst Cost Analysis for Turbines - Uncontrolled Emissions ................ 63
Table 23 - Oxidation Catalyst Cost Analysis for Turbines - BAT Emissions ............................. 63
Table 24 - Combustion Control Device Cost Analysis................................................................. 65
Table 25 - LDAR Costs with ONE Future Equipment Costs and EDF Assumptions.................. 66
Table 26 - LDAR Costs with ONE Future Equipment Costs and Assumptions .......................... 67
Table 27 - Super-Emitter Study Data and LDAR Costs ............................................................... 68
Table 28 - LDAR Costs from Vendor Quotes .............................................................................. 70
Table 29 - LDAR Cost Analysis Based on Rella et. al. and British Columbia Emissions
Assumptions...................................................................................................................... 70
Table 30 - Non-Road Compression Ignition Engine Emission Standards.................................... 73
Table 31 - Non-Road Spark Ignition Engine Emisison Standards ............................................... 73
Figures:
Figure 1: Drawing of a barrel type design of a pig launcher and receiver equipped with
uncontrolled depressurization vents........................................................................................ 42

02/04/2017
Page
5
of
73
Abbreviations and Acronyms:
A/F
Air-to-Fuel
AVO
Auditory, Visual, and Olfactory Inspections
BAT
Best Available Technology
bhp
Brake Horsepower
BMP
Best Management Practices
BTEX
Benzene, Toluene, Ethylbenzene, and Xylene
CFR
Code of Federal Regulations
CO
Carbon Monoxide
CO
2
Carbon Dioxide
CPI
Consumer Price Index
DEA
Diethanolamine
DEP
Pennsylvania Department of Environmental Protection
EGR
Exhaust Gas Recirculation
EPA
U.S. Environmental Protection Agency
g/bhp-h
Grams per Break Horesepower-Hour
GHG
Greenhouse Gas(es)
GP
General Plan Approval/General Operating Permit
GP-5
General Plan Approval/General Operating Permit for Natural Gas Compressor
Stations, Processing Plants, and Transmission Stations
GP-5A
General Plan Approval/General Operating Permit for Unconventional Natural
Gas Well Site Operations and Remote Pigging Stations
H
2
O
Water
H
2
S
Hydrogen Sulfide
HAP
Hazardous Air Pollutant
HCHO
Formaldehyde
GPU
Gas Production Unit
MEA
Monoethanolamine
Mcf
Thousand Cubic Feet
MMBtu
Million British Thermal Units
MMBtu/h
Million British Thermal Units per Hour
N
2
Molecular Nitrogen
NAAQS
National Ambient Air Quality Standard
NESHAP
National Emission Standards for Hazardous Air Pollutants
NGL
Natural Gas Liquids
NGStar
The Natural Gas Star Program
NMNEHC
Non-Methane, Non-Ethane Hydrocarbon
NO
Nitric Oxide
NO
2
Nitrogen Dioxide
NO
X
Oxides of Nitrogen
NSCR
Non-Selective Catalytic Reduction
NSPS
New Source Performance Standards
OGI
Optical Gas Imaging Camera
PennDOT
Pennsylvania Department of Transportation
PM
Particulate Matter
PM
2.5
Particulate Matter with an Aerodynamic Diameter Less Than 2.5 Microns
PM
10
Particulate Matter with an Aerodynamic Diameter Less Than 10 Microns
Ppmdv
Parts Per Million, Dry, by Volume

02/04/2017
Page
6
of
73
ppmv
Parts Per Million by Volume
PRO
Partner Reported Opportunities
PTE
Potential to Emit
RBLC
RACT/BACT/LAER Clearinghouse
REC
Reduced Emission Completion
scf
Standard Cubic Feet
SCR
Selective Catalytic Reduction
SI RICE
Spark Ignition Reciprocating Internal Combustion Engine
SO
2
Sulfur Dioxide
SO
X
Oxides of Sulfur
THC
Total Hydrocarbons
tpy
Tons Per Year
TSD
Technical Support Document
VOC
Volatile Organic Compound

02/04/2017
Page
7
of
73
Executive Summary:
This technical support document (TSD) was prepared by the Pennsylvania Department of
Environmental Protection (Department or DEP) to include background information about
General Plan Approval and General Operating Permit for Unconventional Natural Gas Well Site
Operations and Remote Pigging Stations (GP-5A), and Natural Gas Compressor Stations,
Processing Plants, and Transmission Stations (GP-5) released on
(Insert Date).
The TSD
includes the rationale for the air permitting requirements for air pollution sources and the
associated air pollution controls covered by the General Permits. The TSD describes sources in
common to all types of facilities, sources specific to GP-5A eligible facilities, and sources
specific to GP-5 eligible facilities.
According to 25 Pa. Code §127.1, air contamination sources must be regulated to protect the
public welfare and new sources shall control air pollutant emissions to the maximum extent
consistent with Best Available Technology (BAT) as determined by the Department. Also, the
federal New Source Performance Standards (NSPS) are incorporated into the Department’s
regulations by reference in 25 Pa. Code §122.3, and federal National Emissions Standards for
Hazardous Air Pollutants (NESHAP) regulations are incorporated by reference in 25 Pa. Code
§127.35. The current requirements of U.S. Environmental Protection Agency’s (EPA) NSPS for
Crude Oil and Natural Gas Facilities (40 CFR Part 60, Subparts OOOO and OOOOa) and other
federal requirements are streamlined into the proposed General Permits.
Introduction:
Pursuant to Section 6.1 of the Air Pollution Control Act (APCA, 35 P.S Section 4006.1) and
25 Pa. Code §§127.514 and 127.611, the Department may issue General Plan Approvals and
General Operating Permits (General Permits or GPs) for specific categories of sources that are
similar in design and operation and can be adequately regulated with standardized specifications
and conditions.
The Department first issued GP-5 on March 10, 1997, for natural gas production wells only.
This established BAT for stationary natural gas-fired spark ignition reciprocating internal
combustion engines (engines) rated from 100 brake horsepower (bhp) to 1,500 bhp and for
glycol dehydrators with an uncontrolled potential to emit (PTE) volatile organic compounds
(VOC) in excess of 10 tons per year (tpy) at natural gas production facilities. Then on July 26,
2003, the Department published the Air Quality Permit Exemptions list, which specified sources
or classes of sources that were determined to be exempt from the plan approval and permitting
requirements of the Pennsylvania Air Pollution Control Act (APCA), 35 P.S. §4001
et seq.
and
25 Pa. Code Chapter 127, which included crude oil and natural gas wells. On July 27, 2006, the
DEP revised the GP-5 to expand the applicability to natural gas, coal bed methane, and gob gas
production or recovery facilities including gathering stations. On February 2, 2013, the
Department changed the applicability of GP-5 to natural gas compression and/or processing
facilities, and established BAT requirements for engines, stationary natural gas-fired simple
cycle turbines (turbines), natural gas compressors, storage vessels, glycol dehydrators, natural
gas fractionation units, equipment leaks, pneumatic controllers, and sweetening units. The Air
Quality Permit Exemptions were amended by the Department on August 10, 2013, to change the
unconditional exemption of all crude oil and natural gas wells to the unconditional exemption of
sources located at conventional well sites and a conditional exemption for sources located at
unconventional natural gas well sites.

02/04/2017
Page
8
of
73
Governor Tom Wolf introduced a methane reduction strategy on January 19, 2016. The four-
point plan included the following strategies:
?
To reduce leaks at new unconventional natural gas well pads, DEP will develop a new
general permit for oil and gas exploration, development, and production facilities,
requiring BAT for equipment and processes, better recordkeeping, and quarterly
monitoring inspections.
?
To reduce leaks at new compressor stations and processing facilities, DEP will revise its
current general permit, updating best available technology requirements and applying
more stringent leak detection and repair (LDAR) and other requirements to minimize
leaks.
?
To reduce leaks at existing oil and natural gas facilities, DEP will develop a regulation
for existing sources for consideration by the Environmental Quality Board.
?
To reduce emissions along production, gathering, transmission and distribution lines,
DEP will establish best management practices (BMP), including LDAR programs.
The new GP-5A is applicable to the sources located at unconventional natural gas well site
operations and remote pigging stations, and the revised GP-5 is applicable to sources located at
natural gas compressor stations, processing plants, and transmission stations. The use of the
GP-5 and GP-5A are restricted to facilities with actual emissions less than 100 tpy of criteria
pollutants (NO
X
, CO, SO
2
, PM
10
, and PM
2.5
), less than 50 tpy of VOC, less than 10 tpy of any
single HAP, and less than 25 tpy of total HAPs. For facilities located in Philadelphia, Bucks,
Chester, Montgomery, or Delaware counties, the NO
X
and VOC emissions thresholds are less
than 25 tpy each.
Wherever possible, the terms and conditions of both General Permits have been streamlined to
incorporate both federal and state requirements. It is the duty of the Responsible Official to
ensure that the facility is in compliance with all applicable federal, state, and local laws and
regulations, including 25 Pa. Code, Subpart C, Article III. Nothing in this General Permit
relieves the Responsible Official from this obligation to comply.
Definitions:
Words and terms that are not otherwise defined in this General Permit have the meanings set
forth in Section 3 of the APCA (35 P.S. § 4003) and Title 25, Article III including 25 Pa. Code
§ 121.1 (relating to definitions) unless the context indicates otherwise. The meanings set forth in
applicable definitions codified in the Code of Federal Regulations including 40 CFR Part 60
Subparts JJJJ, KKKK, OOOO, and OOOOa or 40 CFR Part 63 Subparts HH and ZZZZ also
apply to these general permits.
Coal bed methane
– Methane that is extracted from a coal bed and the surrounding rock strata
by extraction wells drilled in advance of a mining operation, which is typically of pipeline
quality.
Deviation
– An instance in which an affected source or the owner or operator of an affected
source fails to meet any term or condition of this General Permit, including any emission limit,

02/04/2017
Page
9
of
73
operating limit, or work practice standard, including during startup, shutdown, or malfunction
regardless of whether such a failure is permitted.
Difficult-to-Monitor
– A fugitive emissions component that cannot be monitored without
elevating the monitoring personnel more than 6.6 feet above the surface may be designated
difficult-to-monitor for the purposes of Section K.
Fugitive Emissions Component
– Any component that has the potential to emit fugitive
emissions of methane, VOC, or HAP at a facility, but no limited to, valves, connectors, pressure
relief devices, open-ended lines, flanges, compressors, instruments, meters, covers, and closed
vent systems. Devices that vent as part of normal operations are not considered fugitive sources
unless the emission originates from a place other than the vent.
Gob gas
– Methane that is mixed with air from a mine ventilation system due to the mining
operation reaching the area of an extraction well, which is typically below pipeline quality.
Haul Road
– A road owned or operated by the permittee which is used to facilitate the
movement of people, equipment, and/or materials to and from a facility.
Leak –
Any release of gaseous hydrocarbons that is detected by Auditory, Visual, and Olfactory
(AVO) inspection; an optical gas imaging (OGI) camera; a gas leak detector that meets the
requirements of 40 CFR Part 60, Appendix A-7, Method 21; or other leak detection methods
approved by the Department’s Division of Source Testing and Monitoring. However, a release
from any equipment or component designed by the manufacturer to protect the equipment,
controller, personnel, or to prevent ground water contamination, gas migration, or an emergency
situation is not considered a leak.
Malfunction
– Any sudden, infrequent, and not reasonable preventable failure of air pollution
control equipment, process equipment, or a process to operate in a normal or usual manner.
Failures caused in part by poor maintenance or careless operation are not malfunctions.
Malfunctions include, but are not limited to, triggering of emergency shutdown devices and
unscheduled blowdowns.
Natural Gas Compressor Station
– A facility that compresses and/or processes natural gas,
coal bed methane, or gob gas prior to the point of custody transfer using processes including, but
not limited to, gas dehydration, compression, pigging, and storage.
Natural Gas Processing Plant
– A facility that engages in the extraction of natural gas liquids
from field gas, the fractionation of mixed natural gas liquids to natural gas products, or both
extracts and fractionates natural gas liquids.
Natural Gas Transmission Station
– A facility that compresses and/or process natural gas after
the point of custody transfer using processes including, but not limited to, gas dehydration,
compression, pigging, and storage.
Pigging Operations
– The process of removing and collecting condensed liquids including
condensate, intermediate hydrocarbons, or produced water from a pipeline using a spherical or
bullet-shaped device, known as a pig, forced through the pipeline by natural gas pressure. The
liquids are then collected at their eventual destination in a storage tank, often referred to as a slug
tank.

02/04/2017
Page
10
of
73
Point of Custody Transfer
– The location after the processing and/or treatment of natural gas in
the production sector, typically after a natural gas processing plant, where control and/or
ownership of the natural gas is transferred from one owner or operator to another.
Remote Pigging Station
– A facility where pigging operations are conducted that is not located
at an unconventional natural gas well site, natural gas compressor station, natural gas processing
plant, or natural gas transmission station and which meets or exceeds the exemption emissions
thresholds of 200 tpy of methane, 2.7 tpy of total VOC, 0.5 tpy of a single HAP, or 1.0 tpy of
total HAP.
Sour Gas
– Natural gas where the H
2
S content is in excess of 4 ppmv at standard temperature
and pressure.
Start of Production
– The beginning of initial flow following the end of flowback when there is
continuous recovery of salable quality gas and separation and recovery of any crude oil,
condensate, or produced water. A well whose owner or operator is selling gas through
temporary equipment designed for flowback shall not be considered in production until either the
sales continue through the temporary equipment for more than 30 days or the gas is routed to a
permanent production separator.
Unconventional Natural Gas Well
- A well drilled to produce natural gas from shale
formations below the Elk Group or its geologic equivalent stratigraphic interval, where recovery
of the resource is generally not economic without the bores being stimulated by hydraulic
fracturing, multilateral well bores, or other techniques to expose more of the formation to the
well bore.
Unconventional Natural Gas Well Site
– A location with one or more unconventional natural
gas wells at which unconventional natural gas well site operations are conducted.
Unconventional Natural Gas Well Site Operations -
Equipment and processes at
unconventional natural gas well sites including, but not limited to, drilling, hydraulic fracturing
or refracturing, well completion, gas dehydration, tanker truck load-out, wellbore liquid
unloading, gas compression, pigging, and storage.
Unsafe-to-Monitor
– A fugitive emissions component that cannot be monitored because
monitoring personnel would be exposed to immediate danger while conducting a monitoring
survey may be designated unsafe-to-monitor for purposes of Section K.
Wellbore Liquids Unloading
– The process of removing accumulated liquids from a natural gas
well in order to restore well pressure and natural gas production.
Well Completion -
The beginning of the flowback period after hydraulic fracturing is
completed. If a well is shut-in prior to the beginning of flowback, the well is not considered to
be completed.
Applicability/Scope:
GP-5A authorizes the construction, modification, and/or operation of sources located at
unconventional natural gas well site operations. The applicability of this general permit may
include one or more of the following operations or emissions sources:

02/04/2017
Page
11
of
73
?
Fugitive Particulate Matter
?
Well Drilling and Hydraulic Fracturing Operations
?
Well Completion Operations
?
Natural Gas-Fired Combustion Units
?
Glycol Dehydration Units
?
Stationary Natural Gas-Fired Spark Ignition Internal Combustion Engines
?
Reciprocating Compressors
?
Storage Vessels
?
Tanker Truck Load-Out Operations
?
Fugitive Emissions Components
?
Controllers
?
Pumps
?
Enclosed Flares and Other Control Devices
?
Pigging Operations
?
Wellbore Liquids Unloading Operations
GP-5 authorizes the construction, modification, and/or operation of sources located at natural gas
compression and/or processing facilities. The applicability of this general permit may include
one or more of the following operations or emissions sources:
?
Fugitive Particulate Matter
?
Natural Gas-Fired Combustion Units
?
Glycol Dehydration Units
?
Stationary Natural Gas-Fired Spark Ignition Internal Combustion Engines
?
Stationary Natural Gas-Fired Combustion Turbines
?
Reciprocating Compressors
?
Centrifugal Compressors
?
Storage Vessels
?
Tanker Truck Load-Out Operations
?
Fugitive Emissions Components
?
Controllers
?
Pumps
?
Enclosed Flares and Other Control Devices
?
Pigging Operations
An Application for Authorization to Use GP-5 or GP-5A may be submitted for the operation of
an eligible source if the source is exempted from plan approval requirements under 25 Pa. Code
§127.14. If any source located at a facility cannot be regulated under the appropriate general
permit, a plan approval and/or an operating permit issued in accordance with 25 Pa. Code,
Chapter 127, Subchapter B and/or Subchapter F will be required.
Prohibited Use of GP-5 and GP-5A:
The proposed GP-5 and GP-5A are different from other General Permits issued by the
Department. The general permit program typically establishes a general plan approval/general
operating permit with requirements for a specific type of
source
, and can be used at Title V
facilities when they are adding that type of source to the facility. However, GP-5 and 5A are
different in that they establish requirements for a specific category of
facility
, which must be

02/04/2017
Page
12
of
73
either a natural or synthetic minor facility. No Title V facility can use the GP-5 or 5A to
construct the sources listed therein.
In addition, because most of the natural gas production in the Marcellus Shale and Utica Shale
Region is not sour gas, the GP-5 and 5A are prohibited for use by those facilities that produce or
process sour gas. This removes the necessity to include the sweetening unit requirements and
means many of the federal SO
2
requirements are presumptively met.
The owner or operator of a facility is also prohibited from circumventing the requirements for
Title V applicability, the prevention of significant deterioration, or non-attainment new source
review by allowing a pattern of ownership or development that conceals that the facility would
otherwise be required to submit a plan approval or operating permit application. This includes
specifically phasing, staging, delaying, or engaging in incremental construction over the
geographical extent of the facility or using a device, stack height that exceeds good engineering
practice, dispersion technique, or other technique to conceal or dilute emissions of air
contaminants without reducing the amount of emissions in order to appear to qualify for the
General Permit when the facility should actually be authorized under a plan approval or
operating permit.
Authorization to Use GP-5 or GP-5A:
In the previous version of GP-5, issued February 2, 2013, there were five conditions related to
the proper use of the General Permit. These conditions were combined into a single condition in
the proposed GP-5A and the modified GP-5 for ease of reference. The intent was to provide a
single reference section that allows the owner or operator to easily locate information about the
General Permit’s lifecycle.
The application requirements were updated to include the General Information Form
(1300-PM-BIT0001). Instructions are now included for an administrative amendment in the
event of a change in the name, address, or telephone number of a person identified in the General
Permit or other similar minor administrative changes.
General Permit Fees:
The condition regarding fees remains as a separate condition for clarity and ease of amendment.
An Administrative Amendment Fee was added to this section. Both fees are referred to in the
section titled Authorization to Use GP-5 or GP-5A.
Applicable Laws:
Wherever possible, the terms and conditions of the proposed GP-5 and 5A have been streamlined
to satisfy both federal and state requirements. It is the duty of the Responsible Official to ensure
that the facility is in compliance with all applicable laws and regulations, and nothing in the
General Permit relieves the Responsible Official of that duty. Therefore, it is suggested that
owners and operators carefully review the listed sources and federal, state, and local
requirements and compare them to the streamlined permit. If this is accomplished during the
comment period, the permit may be edited to ensure that any applicable requirements that were
omitted can be incorporated prior to finalization.

02/04/2017
Page
13
of
73
The Department’s review has found that the applicable federal regulations include the following
NSPS and NESHAP subparts:
(a)
40 CFR Part 60, Subpart JJJJ – Standards of Performance for Stationary Spark
Ignition Internal Combustion Engines.
This subpart establishes emission standards
and compliance requirements for the control of emissions from stationary spark
ignition internal combustion engines that commenced construction, modification or
reconstruction after June 12, 2006, where the SI-RICE are manufactured on or after
specified manufacture trigger dates. The manufacture trigger dates are based on the
engine type, fuel used, and maximum engine horsepower. The state BAT
requirements for engines at natural gas compression and/or processing facilities under
the current GP-5 are either equivalent or more stringent than the federal requirements.
(b)
40 CFR Part 60, Subpart KKKK – Standards of Performance for Stationary
Combustion Turbines.
This subpart establishes emission standards and compliance
schedules for the control of emissions from stationary combustion turbines with a
heat input at peak load equal to or greater than 10 million British thermal units per
hour (MMBtu/h) that commenced construction, modification or reconstruction after
February 18, 2005. The pollutants regulated by this subpart are NO
X
and sulfur
dioxide (SO
2
). However, the SO
2
requirements can be met by fuel composition
analysis, and the prohibition of using the General Permits for facilities that produce or
process sour gas ensures that the fuel composition analysis will be met.
(c)
40 CFR Part 60, Subpart OOOO
Standards of Performance for Crude Oil and
Natural Gas Production, Transmission, and Distribution for which
Construction, Modification, or Reconstruction Commenced after August 23,
2011, and on or before September 18, 2015.
This subpart establishes emission
standards and compliance schedules for the control of VOC and SO
2
emissions from
affected facilities. In Subpart OOOO, the only SO
2
emission source is the sweetening
unit typically located at natural gas processing plants. The prohibition of using the
General Permits for facilities that produce or process sour gas makes it unlikely that a
sweetening unit will produce significant SO
2
emissions, so the requirements for the
sweetening unit were not included.
(d)
40 CFR Part 60, Subpart OOOOa – Standards of Performance for Crude Oil
and Natural Gas Facilities for which Construction, Modification, or
Reconstruction Commenced After September 18, 2015.
This subpart establishes
emission standards and compliance schedules for the control of the pollutant
greenhouse gases (GHG). The GHG standard in this subpart is in the form of a
limitation on emissions of methane. This subpart also establishes emission standards
and compliance schedules for the control of VOC and SO
2
emissions. As for
Subpart OOOO, the prohibition of using the General Permits for facilities that
produce or process sour gas makes it unlikely that a sweetening unit will produce
significant SO
2
emissions and the requirements for the sweetening unit were not
included.
(e)
40 CFR Part 63, Subpart HH – National Emission Standards for Hazardous Air
Pollutants from Oil and Natural Gas Production Facilities.
This subpart applies
to the owners and operators of affected units located at natural gas production
facilities that are major or area sources of HAPs, and that process, upgrade, or store

02/04/2017
Page
14
of
73
natural gas prior to the point of custody transfer, or that process, upgrade, or store
natural gas prior to the point at which natural gas enters the natural gas transmission
and storage source category or is delivered to a final end user. Because the GP-5 and
5A require that the facility is a natural or synthetic minor, only the area source
requirements apply. Area sources are broken down into two categories, those located
within an Urbanized Area (UA) plus offset or an Urbanized Cluster (UC) and those
not located within a UA plus offset or UC.
(f)
40 CFR Part 63, Subpart ZZZZ – National Emission Standards for Hazardous
Air Pollutants for Stationary Reciprocating Internal Combustion Engines
(RICE).
This rule establishes national emission limitations and operating limitations
for HAPs emitted from stationary RICE. This rule applies to owners or operators of
new and reconstructed stationary RICE of any horsepower rating which are located at
a major or area source of HAP emissions. While all stationary RICE located at major
or area sources are subject to the final rule, in the context of these General Permits,
only four-stroke engines above 500 bhp have BACT requirements; all others have
work practice requirements. In addition, for engines that commenced construction,
modification or reconstruction after June 12, 2006, compliance with 40 CFR Part 60,
Subpart JJJJ is compliance with this subpart.
Compliance Requirements and Compliance Certification:
These General Permits function as a plan approval and operating permit only for a synthetic
minor or natural minor facility. The Conditions of Section A lay out general terms and
conditions to ensure that a facility remains a minor facility. The primary requirement to use
these GPs is that the emissions from all sources and associated air pollution control equipment
located at a facility must not exceed the major source thresholds on a 12-month rolling sum
basis.
The owner or operator must constrain the throughput and/or equipment hours of operation to
ensure that the major source thresholds are not exceeded and keep adequate records to
demonstrate compliance. Also, the Responsible Official must sign and submit a Certification of
Compliance with the annual report.
This Condition also contains requirements that all sources and associated air pollution control
equipment are operated so as not to cause air pollution, operated and maintained in accordance
with the manufacturer’s specifications, and operated and maintained to limit the detection of
malodors outside the property boundary. Another key compliance requirement is that the
General Permits cannot be used to relax BAT previously established through the air quality
permitting process.
Notification Requirements:
There are several notifications that the owner or operator of a facility must perform, including a
municipal notification to the local governments where the air pollution source is to be located. A
copy of this notification is required to be included in the Application for Authorization to Use
GP-5 or GP-5A. The notification to the local governments should include a description of the
proposed sources and/or modification of existing sources to be authorized under the application.

02/04/2017
Page
15
of
73
Notifications to the Department include commencement of and completion of construction of a
source authorized under the permit, notice of a lapse of construction activity of 18 months or
more for a source, and prior notification of the commencement of operation of a source.
Notification of malfunctions now includes any unscheduled blowdown or emergency shutdown.
For more information on malfunction reporting, please see the
GP-5 Malfunction Reporting
Instructions
posted on the Department’s website. A requirement to notify the Department prior
to a scheduled blowdown has been added. All notices must be submitted to the Air Program
Manager of the appropriate DEP Regional Office by email to the addresses below:
?
For the Northeast Regional Office:
ER, GP-5A Submittals NERO
?
For the Southeast Regional Office:
ER, GP-5A Submittals SERO
?
For the North Central Regional Office:
ER, GP-5A Submittals NCRO
?
For the South Central Regional Office:
ER, GP-5A Submittals SCRO
?
For the Northwest Regional Office:
ER, GP-5A Submittals NWRO
?
For the Southwest Regional Office:
ER, GP-5A Submittals SWRO
?
For the Philadelphia Air Management Services:
ER, GP-5A Submittals AMS
?
For the Allegheny County Health Department:
ER, GP-5A Submittals ACHD
Recordkeeping Requirements:
Any records generated as part of the terms and conditions of the General Permits are required to
be maintained on site or at the nearest local field office for a minimum of 5 years and may be
maintained in electronic format. The key records generated and maintained by the owner or
operator of a facility authorized under the General Permit are those that show the facility is in
compliance with the facility-wide emission limits on a 12-month rolling basis. All records,
reports, or other information obtained by the Department under the General Permit is publically
available unless the owner or operator of the facility shows cause that the information is
confidential. Under no circumstance are records of emission data eligible for confidentiality.
Reporting Requirements:
In the effort to streamline the federal and state reporting requirements, both were merged into a
single annual report that is required to be sent to EPA and to the Department by March 1
st
of
each year for the previous calendar year. The initial compliance report may cover a period of
less than one year. The report also serves as the basis for the Compliance Certification, which
the Responsible Official must sign. The annual report must be submitted in hard copy to the Air
Program Manager of the appropriate DEP Regional Office and may also be submitted via email
to the appropriate email address listed in the Section on Notification Requirements. Concurrent
to the annual report, the owner or operator of the facility must submit an annual emissions
inventory to the Department by March 1
st
of each year.
Source Testing Requirements:
All submittals related to Source Testing must include two hard copies sent to the Air Program
Manager of the appropriate DEP Regional Office and an electronic copy sent to the appropriate
email address listed in the Section on Notification Requirements.

02/04/2017
Page
16
of
73
The submission of a test protocol must be done at least 60 days prior to the performance of a
source test unless the owner or operator follows the Standardized Performance Test Procedure
outlined in the General Permits. The standardized procedure outlines the conditions and methods
that the Department considers appropriate testing protocol, allowing the owner or operator to
bypass the test protocol submission and use the presumptively approved protocol.
The frequency of the tests required is based on federal and state requirements. In most cases, the
requirements are identical; however, in the case of engines there are some discrepancies in the
timing and frequency of the tests. These discrepancies are outlined in the table below, with the
caveats noted below the table.
Table 1 - Engine Source Testing Requirements
Engine Size
Initial Compliance
Performance Test
Continuous Compliance
Performance Test
Periodic
Monitoring
<100 hp
None Required
None Required
Every 2,500 hours
of operation
100 hp ≤ ER ≤ 500
hp
Within 180 days of
startup of the engine
Within 180 days of each
reauthorization
Every 2,500 hours
of operation
>500 hp
Within 180 days of
startup of the engine
Every 8,760 hours of operation
or
every three years
and
within 180 days of each
reauthorization
Every 2,500 hours
of operation
> 500 hp and
subject to 40 CFR
Part 60, Subpart
ZZZZ
Not Applicable
Every year
Every 2,500 hours
of operation
For an engine greater than or equal to 100 hp and less than or equal to 500 hp, if the engine is
certified by the manufacturer in accordance with 40 CFR Part 60, Subpart JJJJ and the owner or
operator operates and maintains the engine in accordance with the manufacturer’s instructions,
the performance testing requirements are waived.
For an engine greater than 500 hp, if the engine is certified by the manufacturer in accordance
with 40 CFR Part 60, Subpart JJJJ and the owner or operator operates and maintains the engine
in accordance with the manufacturer’s instructions, the continuous compliance performance
testing requirements every 8,760 hours of operation or every three years are waived.
There are also some differences between the federal and state requirements for testing for
combustion turbines, primarily which pollutants are analyzed during the test. The Department
requires testing for NO
X
, CO, NMNEHC (as propane), and total PM while the federal regulation
requires testing of NO
X
and SO
2
only. It is the Department’s determination that SO
2
testing is
not required based on 40 CFR §60.4415(a)(1) if the fuel used is either pipeline quality gas or
field gas that does not meet the definition of sour gas. There is also a frequency difference in
that the federal regulations require annual testing for turbines that do not use water or steam
injection for NO
X
control. The annual requirement can be waived if the owner or operator
installs a continuous monitoring system.
General Methodology of Determining Best Available Technology:
New sources are required to control the emission of air pollutants to the maximum extent,
consistent with BAT as determined by the Department. BAT is defined in 25 Pa. Code §121.1 as

02/04/2017
Page
17
of
73
equipment, devices, methods, or techniques as determined by the Department which will prevent,
reduce, or control emissions of air contaminants to the maximum degree possible and which are
available or may be made available. The applicable emission limits of federal NSPS and
NESHAPs serve as a baseline for determining BAT. The resources utilized in the determination
of BAT include the conditions established for similar sources in Category Number 38 of the Air
Quality Permit Exemption document; the current GP-5; the data in the EPA’s
RACT/BACT/LAER Clearinghouse (RBLC); case-by-case BAT determined in recently issued
plan approvals; and permits recently issued by other states. The Department also evaluated
vendors’ guaranteed emission limits, available stack test data for the applicable sources, and
documents related to EPA’s Natural Gas Star (NGStar) program. The emission limitations
included in the general permits must be technically and economically feasible and must be
sustainable during the life of the air pollution source.
BAT is determined for each pollutant emitted by each source. The general classes of air
emissions from the sources at facilities covered by GP-5 and GP-5A include NO
X
, CO, VOC,
HAP, SO
X
, PM
10
, PM
2.5
, CO
2
, and methane. These pollutants are described in more detail
below.
Oxides of Nitrogen
Oxides of nitrogen are a family of compounds which include nitric oxide (NO) and nitrogen
dioxide (NO
2
) that are produced as a byproduct of the combustion of fuel and air. The heat of
combustion causes the molecular nitrogen (N
2
) in the combustion air to disassociate and oxidize,
forming NO and NO
2
. NO
X
is a criteria pollutant, but it is also a precursor to acid rain, ozone,
and PM
2.5
; NO
X
emissions are typically expressed as NO
2
.
There are three types of NO
X
created during combustion: thermal, fuel, and prompt. Thermal
NO
X
is produced at very high temperatures by the reaction of atmospheric oxygen and nitrogen
and is heavily influenced by combustion temperature. Fuel NO
X
results from oxidation of
nitrogen contained in the fuel. Prompt NO
X
is formed from molecular nitrogen in the air
combining with fuel in fuel-rich conditions.
Strategies for the control of NO
X
include combustion control and post-combustion control. In
combustion control, the combustion temperature is lowered in order to limit the disassociation of
molecular nitrogen through premixing, staging, or excess air. Post-combustion control typically
includes an add-on device, such as nonselective catalytic reduction systems (NSCR) or selective
catalytic reduction (SCR) systems. Both of these systems are described in greater detail in the
individual source BAT analysis.
Carbon Monoxide
Carbon monoxide is a colorless, odorless gas that results from the incomplete combustion of
carbon. CO is formed when insufficient oxygen or poor mixing interferes with the combustion
reaction to produce CO
2
. CO formation is greatest when the air-fuel mixture is rich; however,
CO also forms when a very fuel-lean mixture cannot sustain complete combustion.
Techniques for control of carbon monoxide also include combustion control and post-
combustion controls. As for combustion control, a balance must be sought with NO
X
control as
lower combustion temperatures that prevent thermal NO
X
formation can lead to incomplete
combustion and therefore increase CO formation. Post-combustion controls include nonselective

02/04/2017
Page
18
of
73
catalytic reduction systems and oxidation catalysts. Individual source BAT analyses tend to
favor minimizing NO
X
production through combustion control techniques even though there
may be an increase of CO, and then using post-combustion controls to mitigate the CO impact,
due to Pennsylvania’s status as an Ozone Transport Region non-attainment state.
Volatile Organic Compounds
VOCs are any compounds of carbon, excluding those listed in 40 CFR §51.100(s). The listed
compounds, which include methane and ethane, are excluded because they have been determined
to have negligible photochemical reactivity in the production of photochemical smog. Methane
and ethane are excluded from VOC regulations and measurements for this reason. VOCs are
emitted from combustion sources as unburned fuel. VOCs are also emitted through fugitive
emissions, venting, and the processing of natural gas. For SI-RICE and turbines, the Department
uses non-methane, non-ethane hydrocarbons (NMNEHC) in lieu of VOC. Also, the Department
decided to have a separate emission limit for formaldehyde for engines, which is both a VOC
and a HAP. Therefore, engines will have an emission limit for NMNEHC, which excludes
formaldehyde, and is expressed as propane. Turbines will have an emission limit for NMNEHC
expressed as propane.
Techniques for the control of VOC are different depending on whether they are byproducts of
combustion, due to venting or processing natural gas, or a result of fugitive emissions. Control
techniques considered for the reduction of post-combustion VOC include NSCR systems and
oxidation catalysts. The primary control technique considered for the reduction of venting or
process emissions is the installation of a closed vent system. The primary control technique
considered for the reduction of fugitive emissions is an LDAR program.
Hazardous Air Pollutants
HAPs are air pollutants known to cause cancer or to have other serious health impacts. There are
currently 187 listed HAPs. While combustion accounts for a portion of HAP emissions, HAPs
are also released through fugitive emissions, venting, and the processing of natural gas. The
HAPs of primary concern at unconventional natural gas well site operations, remote pigging
stations, natural gas compressor stations, processing plants, and transmission stations are n-
hexane; benzene, toluene, ethylbenzene, xylenes (collectively known as BTEX); and
formaldehyde. Formaldehyde (HCHO) is of particular interest because it is the predominant
HAP component of combustion emissions, resulting from the incomplete combustion of
methane.
Similar to the control of VOC, the techniques for mitigating HAP emissions are dependent on
whether the emissions are byproducts of combustion, due to venting or processing natural gas, or
a result of fugitive emissions. The techniques to reduce HAP in these cases are the same as those
used to reduce VOCs.
Oxides of Sulfur
Oxides of sulfur are the byproduct of combustion of a fuel that contains sulfur. In the Marcellus
Shale region, natural gas does not contain sulfur above trace amounts. The Department has
determined that for a typical combustion process using natural gas, SO
2
emissions are of minor
significance. Therefore, neither GP-5 nor GP-5A includes SO
2
emission limitations or SO
2
stack

02/04/2017
Page
19
of
73
testing requirements for combustion sources. An example of this determination of minor
significance is given in the calculation below.
瘃̀䠇㸇 긃 䤇䬇䠇̀⨇
猃爃
̀䠇㸇 긃 䤇䬇䠇̀䨇㴇倇儇万㴇䠇̀䌇㴇伇Ⰾ
䠇㸇 긃 䤇䬇䠇̀㔇ㄇ
䠇㸇 긃 䤇䬇䠇̀⨇
Ⰾ Ⰾ
砃瘃爃礃̀䠇㸇̀㔇ㄇ
䠇㸇 긃 䤇䬇䠇̀㔇ㄇ
Ⰾ Ⰾ
䠇㸇 긃 䤇䬇䠇̀䨇㴇倇儇万㴇䠇̀䌇㴇伇
甃稃砃稃̀伇㼇䈇̀䨇㴇倇儇万㴇䠇̀䌇㴇伇
Ⰾ Ⰾ
伇㼇䈇̀䨇㴇倇儇万㴇䠇̀䌇㴇伇
猃爃甃爃̀␇倇儇
Ⰾ 搒
猃爃
̀␇倇儇
⼇⼇␇倇儇
䰍 砃瘃球 䠍 猃爃
㼋㠋
䠇㸇̀㔇ㄇ
⼇⼇␇倇儇
伍 砃爃爃 䠍 猃爃
㼋㘋
̀
䠇㸇̀㔇ㄇ
⼇⼇␇倇儇
欍瘃爃̀┇⠇㐇̀ᨂ砃爃瘃甃甃爃㨒㴇㬒㨒球㬒漍
Another example is the sweetening unit sulfur feed rate, where the lowest applicable limit is two
long tons per day. Using the equations provided in 40 CFR §60.5406a(b)(1) and assuming the
4 ppm H
2
S limit, the “acid gas” flow rate would have to be nearly 13,500 MMscf/day to reach a
sulfur feed rate of two long tons of sulfur per day. According to an Energy Information
Administration report,
1
an average transmission station moves approximately 700 MMscf/day
and the largest compressor station at the time moves as much as 4,600 MMscf/day. Neither type
of facility would reach the required sulfur feed rate for control under 40 CFR §60.5405(a). This
is also true for gathering stations as they are generally smaller than transmission stations.
Particulate Matter
There are many types of particulate matter emissions, with classifications based on size (i.e.,
PM
10
and PM
2.5
) and state (i.e., filterable and condensable). Some particles are emitted directly
from a source, such as construction sites, unpaved roads, or combustion, and are called primary
particles. Others are formed in complicated reactions in the atmosphere from SO
2
and NO
X
, and
are called secondary particles. The clearing, grading, and construction of a site, as well as the
drilling of a well can create primary particle emissions. The combustion of natural gas produces
very little primary particle emissions, and the control of precursor emissions helps reduce
secondary particle emissions. Because the primary particle emissions from combustion sources
are of minor significance, and the primary precursor emissions are either of minor significance or
well controlled, GP-5 and GP-5A do not include PM emission limitations or stack testing
requirements from most combustion sources.
2
Carbon Dioxide
Carbon dioxide is sometimes present in natural gas in significant quantities and is also a primary
byproduct of combustion. CO
2
is a greenhouse gas; however, there are currently no limitations
in effect for minor sources. Therefore, the explicit control of CO
2
is not proposed for any source
in GP-5 or GP-5A, as no method of explicit control exists except for carbon capture and
sequestration (CCS). CCS is not a viable strategy for minor sources, as it is not cost effective on
that scale. However, many of the BMP, maintenance requirements, and operating limitations
included in the General Permit have the co-benefit of increasing the source’s efficiency and
therefore reducing CO
2
on an output basis.
1
Technical Report: Natural Gas Compressor Stations on the Interstate Pipeline Network: Developments Since 1996,
November 2007, Energy Information Administration, Office of Oil and Gas.
2
Technical Report: Development of Fine Particulate Emission Factors and Speciation Profiles for Oil and Gas-fired
Combustion Systems, Update: Critical Review of Source Sampling and Analysis Methodologies for Characterizing
Organic Aerosol and Fine Particulate Source Emission Profiles, February 2004, Gas Research Institute, et. al.

02/04/2017
Page
20
of
73
Methane
Methane is the primary component of natural gas and represents a major portion of the emissions
from unconventional natural gas well site operations, remote pigging stations, natural gas
compressor stations, processing plants, and transmission stations. While methane is harmless in
low concentrations, it can explode when the concentration reaches the lower explosive limit.
Methane is a greenhouse gas that is 25 times more potent than CO
2
on a ton-for-ton basis. For
these reasons, the Department is taking steps to minimize methane emissions from natural gas
operations.
To this end, the Department developed an exemption threshold by back-calculating a methane
limit using an average natural gas composition of 88.78% methane, 1.25% VOC, and the VOC
exemption threshold of 2.7 tpy as shown in Appendix A. Therefore, a methane exemption
threshold of 200 tpy is established for new or modified unconventional natural gas facilities.
BAT control measures for methane must be implemented above that limit. Even when not
explicitly stated for control, such as for sources that were constructed prior to (
effective date of
the GP-5 and 5A
), the BMP, maintenance requirements, operating limitations, and control
requirements often have a co-benefit of reducing methane emissions.
As an illustration of this co-benefit, assume the 95% control requirement for VOC from a storage
vessel authorized under Category Number 38 is met using an enclosed flare. Assuming a pre-
control VOC emission rate of 10 tpy, an average gas composition as determined in Appendix A,
a methane destruction efficiency of 95%, and that 100% of the destroyed methane becomes CO
2
,
it is possible to calculate the greenhouse gas reduction in CO
2e
. The following equation shows
the amount of methane emissions that coincide with the 10 tpy VOC assumption:
猃爃̀阀鄀退̀ᤀሀ؀
鬀蜀茀鐀
ⰎⰎ
̀猃̀阀鄀退̀退茀阀需鐀茀踀̀褀茀销
爃爃猃稃球̀阀鄀退̀ᤀሀ؀
ⰎⰎ
爃稃稃礃稃̀阀鄀退̀؀଀
猃̀阀鄀退̀退茀阀需鐀茀踀̀褀茀销
Ⰾ 䰍 ̀瘃稃礃稃̀阀鈀鬀̀؀଀
The next calculation shows the amount of CO
2
generated by combusting the methane in the
control device, using the 95% destruction efficiency and 100% conversion assumptions:
瘃稃礃稃̀阀鄀退销̀؀଀
鬀蜀茀鐀
Ⰾ Ⰾ
球爃爃爃̀踀萀̀؀଀
猃̀阀鄀退̀؀଀
ⰎⰎ
猃̀踀萀 긃 輀鄀踀̀؀଀
猃砃爃瘃̀踀萀̀؀଀
ⰎⰎ
爃笃眃̀踀萀 긃 輀鄀踀̀؀଀
猃̀踀萀 긃 輀鄀踀̀؀଀
Ⰾ Ⰾ
猃̀踀萀 긃 輀鄀踀̀؀ሀ
猃̀踀萀 긃 輀鄀踀̀؀଀
ⰎⰎ
瘃瘃爃球̀踀萀̀؀ሀ
踀萀 긃 輀鄀踀̀؀ሀ
Ⰾ Ⰾ
猃̀阀鄀退̀؀ሀ
球爃爃爃̀踀萀̀؀ሀ
䰍 猃球礃猃稃̀阀鈀鬀̀؀ሀ
̀鈀鐀鄀蘀需蔀蜀蘀̀蠀鐀鄀輀̀阀言蜀̀蔀鄀輀萀需销阀謀鄀退̀鄀蠀̀輀蜀阀言茀退蜀
The final equation shows the total amount of CO
2e
reduced by showing the amount of CO
2e
equivalent to the reduced methane and subtracting the CO
2e
of the CO
2
emissions produced in
the combustion process:
㨒爃笃眃㬒㨒瘃稃礃稃̀阀鈀鬀̀؀଀
球眃̀阀鈀鬀̀؀ሀ
㘋挋
猃̀阀鈀鬀̀؀଀
瀍 䘍
㨒猃球礃猃稃̀阀鈀鬀̀؀ሀ
猃̀阀鈀鬀̀؀ሀ
㘋挋
猃̀阀鈀鬀̀؀ሀ
䰍 猃爃甃猃甃眃̀阀鈀鬀̀؀ሀ
㘋挋
̀蜀踀謀輀謀退茀阀蜀蘀̀萀鬀̀阀言蜀̀蔀鄀輀萀需销阀謀鄀退̀鄀蠀̀輀蜀阀言茀退蜀
Techniques for the control of methane emissions differ depending on whether they are due to
venting, processing natural gas, or a result of fugitive emissions. The primary control technique
considered for the reduction of venting or process emissions is the installation of a closed vent
system. The primary control technique considered for the reduction of fugitive emissions is an
LDAR program.

02/04/2017
Page
21
of
73
Sources Common to GP-5 and GP-5A:
Fugitive Particulate Matter
Prior to a facility being constructed, the site must be prepared, which can include, but is not
limited to, clearing, grading, and construction. Activities that have the potential to emit fugitive
dust to the outdoor atmosphere must be controlled in accordance with 25 Pa. Code §§123.1 and
123.2. In addition, unconventional natural gas well sites have the potential for fugitive dust
emissions due to high volumes of heavy truck traffic during the hydraulic fracturing stage. In
order to minimize emissions due to heavy truck traffic, the Department proposes that the
following conditions represent BAT:
?
Preventing emissions that are visible at the point they move outside the property
boundaries and the tracking of dirt or soils onto public roads by implementing
measures including, but not limited to, sweeping and/or the use of a tire washing
system.
?
Promptly removing earth or other material that is deposited by trucking or other
means on public roadways.
?
Applying water or other chemical dust suppressants as needed to haul roads, the
shoulders of access roadways, and the shoulder of the public highway for a
distance of 500 feet in both directions to reduce fugitive dusts based on daily site
conditions.
?
The application of dust suppressants on the public highway shall be done in
accordance with the appropriate PennDOT Bulletins.
?
If water is used, it shall not be applied if the result would be a potentially unsafe
condition.
?
Waste oil or wastewater shall not be used as a dust suppressant.
?
Posting signage consistent with PennDot regulations on haul roads before the
commencement of facility construction imposing a 15 mph speed limit.
?
Posting signage that complies with 67 Pa. Code §212.101(a) and (b) informing
drivers of diesel-powered motor vehicles of the Pennsylvania anti-idling law,
limiting idling to no more than 5 minutes in any continuous 60-minute period
except for the exemptions and exclusions of 35 P.S. §4603(b) and (c).
?
A written procedures document that describes the activities utilized at the facility
to control fugitive particulate matter emissions shall be maintained on-site. The
company shall keep sufficient records to demonstrate that the activities utilized at
the facility to control fugitive particulate matter emissions are being implemented.
These conditions are consistent with the BAT for fugitive particulate matter control from GP-12
Fugitive Dust Sources and Diesel-Fired Internal Combustion (IC) Engines at Coal and Coal
Refuse Preparation Plants.
Natural Gas-Fired Combustion Units
There are many different combustion units used at natural gas production and processing
facilities. Some are small integrated units, typically rated at less than 2.5 MMBtu/h, such as

02/04/2017
Page
22
of
73
those found on gas production units (GPU), heated flash separators, or glycol dehydration units.
Others are large units, some larger than 10 MMBtu/h, such as the fractionation column heaters
found at natural gas processing plants or evaporators found at unconventional natural gas well
sites.
Often when natural gas first exits the wellbore, it contains free water, condensate, and water
vapor that must be removed from the natural gas stream. GPUs perform this task, and many
have small boilers that facilitate the removal of natural gas from the liquids stream through
flashing, which volatilizes the gas from the liquids. In dry-gas regions, the liquid is primarily
water, and is referred to as produced water. In rich-gas regions, the high percentage of
condensate in the natural gas stream often requires further processing to ensure that the water
and condensate are separated. Heated flash separator units are used for this purpose and are also
equipped with small boilers to facilitate the condensate removal from the water by flashing. The
flow of the natural gas and liquids through the GPUs and heated flash separator units are often
controlled by integrated controllers and pumps.
Combustion units with a rated capacity of less than 10 MMBtu/h of heat input fired on natural
gas supplied by a public utility are exempt from plan approval and operating permit requirements
by the Air Quality Permit Exemption document. Under Category Number 39 of the proposed
Air Quality Permit Exemption document, combustion units rated at less than 10 MMBtu/h firing
natural gas supplied by an independent producer shall be exempt from plan approval, and the
GPs may function as the required operating permit. Even though the combustion units are
exempt from plan approval and/or operating permits, the owner or operator will be required to
list these sources in the Application for Authorization to Use GP-5 or GP-5A for reference
purposes to ensure compliance with the facility emissions limits. Any associated fugitive
emissions components, controllers, and pumps will be subject to their respective requirements of
the General Permit.
For combustion units between 10 MMBtu/h and 50 MMBtu/h the BAT limits from the proposed
GP-1 are incorporated into this General Permit. The following table gives the applicable
emission limitations:
Table 2 - BAT Emission Limits for Natural Gas-Fired Combustion Units
Constructed After:
NO
X
(ppmdv
@ 3% O
2
)
CO
(ppmdv
@ 3% O
2
)
PM
(lb/MMBtu)
Opacity
(No more than 3
minutes in an hour)
Opacity
(At any
time)
December 2, 1995
30
300
0.4
20%
60%
(effective date of GP-5)
15
130
0.4
10%
30%
The above emission limitations were based on permitted limits from recently issued plan
approvals. In addition, the owner or operator of a combustion unit of this category shall conduct
an annual inspection and tune-up and either a performance test or an electro-chemical cell
portable analyzer check within 180 days of initial startup and reauthorization of the General
Permit. The owners and operators of any category of combustion unit will be required to meet
notification, recordkeeping, and reporting requirements to track compliance.

02/04/2017
Page
23
of
73
Stationary Natural Gas-Fired Spark Ignition Internal Combustion Engines
In an engine, a mixture of air and fuel is burned within the engine cylinder and the energy of
expanding gases is converted into mechanical work at the engine crank shaft. The relative
proportions of air and fuel in the combusted mixture is called the air-to-fuel (A/F) ratio. The
A/F ratio is called "stoichiometric" if the mixture contains the precise amount of air that supplies
sufficient oxygen for complete combustion of the fuel with no oxygen or fuel left over after
combustion.
Reciprocating engines are grouped into two general categories based on the combustion model
used in their design: “rich-burn” and “lean-burn”. The primary distinction between the two is the
amount of excess air admitted prior to combustion. Rich-burn engines operate with a minimum
amount of air required for combustion and lean-burn engines use 50% to 100% more air than is
necessary for combustion.
In the natural gas industry, engines are used as prime movers to drive compressors or vapor
recovery units and as electric generators. Both rich-burn and lean-burn engines are used in the
natural gas industry.
Emissions from Lean-Burn and Rich-Burn Engines
The main pollutants emitted from the exhaust of SI-RICE are NO
X
, CO, NMNEHC,
formaldehyde, SO
X
, PM, and methane, depending on the composition of the fuel used. Natural
gas is the primary fuel used by sources at unconventional natural gas well sites, remote pigging
stations, natural gas compressor stations, processing plants, and transmission stations. For
engines, natural gas is the only fuel authorized by GP-5A and GP-5.
Emission Control Technology
Control technologies that may be used on engines primarily fall into two categories, combustion
control and post-combustion control.
Combustion Control
Control of combustion temperature is the principal focus of combustion process control in
natural gas-fired engines. Combustion control requires tradeoffs – higher temperatures favor
complete consumption of the fuel and lower residual hydrocarbons and CO but result in
increased NO
X
formation. Lean combustion dilutes the fuel mixture and reduces combustion
temperatures and therefore reduces NO
X
formation. This allows a higher compression ratio or
peak firing pressures resulting in higher efficiency. However, if the mixture is too lean,
misfiring and incomplete combustion may occur, increasing CO and VOC emissions.
3
Because the NO
X
produced by SI-RICE is primarily thermal NO
X
, reducing the combustion
temperature will result in less NO
X
production. Thus, the main strategy for combustion control
is to control the combustion temperature. This is most easily done by adding more air than what
is required for complete combustion of the fuel. This raises the heat capacity of the gases in the
3
Technical Report: Technology Characterization: Reciprocating Engines, March 2015, Prepared by Darrow, K. et.
al. of ICF International on behalf of the EPA and US DOE.

02/04/2017
Page
24
of
73
cylinder so that for a given amount of energy released in the combustion reaction, the maximum
temperature will be reduced.
Combustion temperature can also be controlled to some extent in reciprocating engines by one or
more of the following techniques:
?
Diluting the fuel-air mixture with exhaust gas recirculation (EGR), which replaces
some of the air and contains water vapor that has a relatively high heat capacity
and absorbs some of the heat of combustion.
?
Modifying valve timing, compression ratio, turbocharging, and the combustion
chamber configuration.
Post-Combustion Emission Reduction Technology for Rich-Burn Engines
Three-Way Catalyst (for NO
X
, CO and NMNEHC reduction)
In rich-burn engines, an after-treatment system such as a three-way catalyst, also known as non-
selective catalytic reduction (NSCR), can be added to reduce NO
X
emission levels. Three-way
catalysts use oxygen to treat exhaust emissions. However, three-way catalysts do not use
unburned combustion oxygen to reduce emissions. They make use of the oxygen within the
constituent compounds. Oxygen from NO
X
is used to oxidize the CO and NMNEHC. This
converts the three pollutants into N
2
, CO
2
and H
2
O. Catalysts may be used in series to obtain
lower emission levels. Typically, the reduction level for NO
X
is > 95%, CO is >95%, and
NMNEHC is >50%. For this analysis, the Department has determined that an NSCR is
economically feasible for engines if the cost per ton of NO
X
, CO, and NMNEHC removal is
approximately $5,000; see Appendix C.
Post-Combustion Emission Reduction Technology for Lean-Burn Engines
Oxidation Catalyst (for CO and NMNEHC reduction)
On lean-burn engines, oxidation catalysts using platinum and palladium are effective for
lowering CO and NMNEHC levels in exhaust emissions. Methane is difficult to oxidize at
exhaust temperatures provided by lean-burn engines; therefore, the control efficiency for
methane can be very low. No A/F ratio control system is required with this type of catalyst and
it can be applied to either rich-burn or lean-burn engines. For this analysis, the Department has
determined that an oxidation catalyst is economically feasible for engines if the cost per ton of
CO and NMNEHC removal is approximately $5,000; see Appendix C.
Selective Catalytic Reduction (for NO
X
reduction)
Selective Catalytic Reduction (SCR) is an exhaust gas after-treatment that specifically targets the
NO
X
in engine exhaust and converts it to N
2
and H
2
O. Unlike the three-way catalyst which uses
oxygen from the exhaust stream to treat emissions, SCR injects a compound into the exhaust
stream to start the reaction. The process begins when a small amount of urea is injected into the
exhaust stream. After hydrolysis, the urea becomes ammonia and reacts with NO
X
. On closed-

02/04/2017
Page
25
of
73
loop control systems SCR can reduce natural gas-fired engine NO
X
emissions by 90% as per the
SCR system manufacturers.
4
In the GP-5 issued on February 2, 2013, the Department required add-on control for CO
emissions which also controls VOC and HCHO emissions. The Department also established an
uncontrolled NO
X
emission limit of 0.5 grams per brake horsepower-hour (g/bhp-h) for lean-
burn engines rated at greater than 500 brake horsepower (bhp) based on vendor’s guaranteed
emission rates. The Department evaluated the economic feasibility of SCR for engines and
determined that SCR is economically cost-prohibitive for engines rated below 4,000 bhp. The
Department did not evaluate SCR for larger engines at that time because the information
received from natural gas compression facility owners led us to believe that the typical engine
sizes were between 1,300 bhp and 4,000 bhp.
The Department has reviewed the manufacturers’ current guaranteed level for NO
X
emissions for
lean-burn engines and found that some manufacturers now offer a guaranteed emissions rate of
0.30 g/bhp-h NO
X
for certain engines rated greater than or equal to 1,875 bhp. Our evaluation of
stack test data for engines permitted under previous versions of GP-5 shows that engines rated
greater than or equal to 1,875 bhp can achieve 0.35 g/bhp-h or less for uncontrolled NO
X
emissions in approximately 33% of cases. For the current analysis, the Department has
determined that SCR is economically feasible for engines if the cost per ton of NO
X
removal is
approximately $10,000; see Appendix B.
Engine Size Grouping
The Department chose the engine size groups based on information for various engine makes and
models available; the general permits group the engines into the following categories:
?
Less than 100 bhp;
?
Greater than or equal to 100 bhp and less than 500 bhp;
?
Greater than or equal to 500 bhp and less than 1,875 bhp; and
?
Greater than or equal to 1,875 bhp.
The groupings are comparable to the bhp categories in 40 CFR Part 60, Subpart JJJJ.
Engine Emission Limits
The BAT for most of the engine size categories in the proposed GP-5A and GP-5 is adapted
from the current GP-5 for Natural Gas Compression and/or Processing Facilities, revised
January 16, 2015. For engines rated above a certain size, the Department evaluated uncontrolled
emissions, control efficiency of various controls and associated costs, and stack test results for
SI-RICE to establish BAT.
It should be noted that there were engines permitted through GP-5 prior to February 2, 2013
which could include sources at natural gas well sites. The scope of the revised GP-5, issued on
February 2, 2013, did not include sources at natural gas well sites. Therefore, specific categories
for these existing engines authorized to operate under previous versions of GP-5 and their
emissions limits have been included in GP-5A.
4
Confidential emails from Vendor A and Vendor B, as referenced in Appendix B of this document.

02/04/2017
Page
26
of
73
Rich-Burn Engines Less Than 100 bhp
Vendor data for rich-burn engines less than 100 bhp indicate uncontrolled weighted average
emission rates
5
of 12.9 g/bhp-h for NO
X
, 12.4 g/bhp-h for CO, and 1.9 g/bhp-h for total
hydrocarbons (THC). In order to determine the VOC emission rate, the Department assumed
that 90% of THC is methane and ethane. Doing so results in a VOC/NMNEHC weighted
average emission rate of 0.2 g/bhp-h. Both the weighted average emission rates and BAT
emission rates were used in the BAT analysis.
All NSCR cost estimations were based on an analysis by E
C
/
R
Incorporated
6
, where they
determined total annual costs based on vendor data. The equations are for retrofitted technology
and give the cost in 2009 dollars. The total annual cost in the cost analysis was then multiplied
by the consumer price index (CPI) of 1.12 for inflating 2009 dollars to 2016 dollars. Because the
equations were designed as a retrofit, it is assumed that it is a conservative cost estimate for new
installations. It was assumed that the control efficiencies for NO
X
and CO are 95% and for
NMNEHC is 50%.
Using the weighted average and BAT emission rates, the assumed control efficiencies, and
assuming full-year operation, the control cost for rich-burn engines less than 100 bhp is
estimated between $569 and $8,414 per ton of pollutants reduced. The $8,414 per ton of
pollutants reduced figure is for a 25 hp engine operating at the BAT rate. This is unrealistic and
not representative of uncontrolled emissions from a 25 hp engine, and therefore is considered
anomalous in the BAT determination. Discarding this anomaly, the control cost for an NSCR for
rich-burn engines less than 100 bhp is then estimated to be between $569 and $4,294 per ton of
pollutant reduced. Therefore, the Department determines NSCR to be technically and
economically feasible option and therefore is BAT for rich-burn engines less than 100 bhp with
emission limits of 0.60 g/bhp-h for NO
X
, 0.60 g/bhp-h for CO, and 0.10 g/bhp-h for NMNEHC
based on the weighted average emission factor and the given control efficiencies.
Rich-Burn Engines Greater Than or Equal To 100 bhp but Less Than 500 bhp
Vendor data for rich-burn engines greater than or equal to 100 bhp but less than 500 bhp indicate
weighted average emission rates of 15.9 g/bhp-h for NO
X
, 8.3 g/bhp-h for CO, 1.5 g/bhp-h for
THC, and 0.3 g/bhp-h for NMNEHC. Both the weighted average emission rates and BAT
emission rates were used in the BAT analysis.
Using the weighted average emission rates and the BAT emission rates, the assumed control
efficiencies, and assuming full-year operation, the control cost for rich-burn engines greater than
or equal to 100 bhp but less than 500 bhp is estimated between $138 and $11,480 per ton of
pollutants reduced. The $11,480 per ton of pollutants reduced figure is for a 100 hp engine
operating at the BAT emissions rate, which presumes installation of NSCR. Discarding this
point, the control cost for an NSCR is then estimated to be between $138 and $5,126 per ton of
pollutant reduced. Therefore, the Department determines NSCR to be BAT for rich-burn
engines greater than or equal to 100 bhp but less than 500 bhp where emission limits of
5
The weighted average emission rates for all engine categories is based on data gathered by the Department and
included in the Excel Spreadsheet titled GP-5 NG Engine Data 12292016.
6
Memorandum on Control Costs for Existing Stationary SI RICE, E
C
/
R
Incorporated, June 29, 2010.

02/04/2017
Page
27
of
73
0.25 g/bhp-h for NO
X
, 0.30 g/bhp-h for CO, and 0.20 g/bhp-h for NMNEHC remains as in the
previously issued GP-5.
Rich-Burn Engines Greater Than or Equal To 500 bhp
Vendor data for rich-burn engines greater than or equal to 500 bhp indicate uncontrolled
weighted average emission rates of 15.4 g/bhp-h for NO
X
, 8.2 g/bhp-h for CO, 1.7 g/bhp-h for
THC, and 0.3 g/bhp-h for NMNEHC. Both the weighted average emission rates and BAT
emission rates were used in the BAT analysis.
Using the weighted average emission rates and the BAT emissions rates, the assumed control
efficiencies, and assuming full-year operation, the control cost for rich-burn engines greater than
or equal to 500 bhp is estimated between $30 and $3,256 per ton of pollutants reduced.
Therefore, the Department determines NSCR to be BAT for rich-burn engines greater than or
equal 500 bhp where emission limits of 0.20 g/bhp-h for NO
X
, 0.30 g/bhp-h for CO, and
0.20 g/bhp-h for NMNEHC remains as in the previously issued GP-5.
Lean-Burn Engines Less Than 100 bhp
Vendor data for lean-burn engines less than 100 bhp indicate uncontrolled weighted average
emission rates of 3.2 g/bhp-h for NO
X
, 82.8 g/bhp-h for CO, and 1.6 g/bhp-h for THC. In order
to determine the VOC emission rate, the Department assumed that 90% of THC is methane and
ethane. Doing so results in a NMNEHC weighted average emission rate of 0.2 g/bhp-h.
SCR cost estimations were based on vendor quotes, cited as from Vendor A, Vendor B, and
Vendor C. It was assumed that the control efficiency for SCR is 90% for NO
X
. All oxidation
catalyst cost estimations were based on an analysis by E
C
/
R
Incorporated, where they determined
total annual costs based on vendor data. The equations are for retrofitted technology and give
the cost in 2009 dollars. The total annual cost in the cost analysis was then multiplied by the CPI
of 1.12 for inflating 2009 dollars to 2016 dollars. Because the equations were designed as a
retrofit, it is assumed that it is a conservative cost estimate for new installations. It was assumed
that the control efficiencies for oxidation catalysts are 90% for CO and 50% for NMNEHC. See
Appendices B and C for the analyses.
Using the weighted average emission rates and BAT emission rates, the assumed control
efficiencies, and assuming full-year operation, the control cost for an oxidation catalyst for lean-
burn engines less than 100 bhp is estimated between $111 and $8,523 per ton of pollutants
reduced. The $8,253 per ton of pollutants reduced figure is for a 25 hp engine operating at the
BAT rate. This is unrealistic and not representative of uncontrolled emissions from a 25 hp
engine, and therefore is considered anomalous in the BAT determination. Discarding this
anomaly, the control cost for an oxidation catalyst for lean-burn engines less than 100 bhp is then
estimated to be between $111 and $4,317 per ton of pollutants reduced. Therefore, the
Department determines oxidation catalyst to be BAT for lean-burn engines less than 100 bhp
where an emission limit of 2.00 g/bhp-h CO remains as in the previously issued GP-5 and an
emission limit of 0.20 g/bhp-h for NMNEHC is established based on the weighted average
emission factor and given control efficiencies.
Using the BAT emission rates, the assumed control efficiencies, and assuming full-year
operation, the control cost for SCR for lean-burn engines less than 100 bhp is estimated between
$124,769 and $248,427 per ton of NO
X
reduced. The Department determines SCR is not BAT

02/04/2017
Page
28
of
73
for lean-burn engines less than 100 bhp because it is not economically feasible, and an emission
limit of 1.00 g/bhp-h of NO
X
is established as required under 40 CFR Part 60 Subpart JJJJ.
Lean-Burn Engines Greater Than or Equal To 100 bhp but Less Than 500 bhp
Vendor data for lean-burn engines greater than or equal to 100 bhp but less than 500 bhp indicate
uncontrolled weighted average emission rates of 7.8 g/bhp-h for NO
X
, 6.7 g/bhp-h for CO,
1.9 g/bhp-h for THC, and 0.6 g/bhp-h for NMNEHC.
Using the weighted average emission rates and BAT emission rates, the assumed control
efficiencies, and assuming full-year operation, the control cost for an oxidation catalyst for lean-
burn engines greater than or equal to 100 bhp but less than 500 bhp is estimated between
$286 and $1,956 per ton of pollutants reduced. Therefore, the Department determines oxidation
catalyst to be BAT for lean-burn engines greater than 100 bhp but less than 500 bhp with
emission limits of 0.70 g/bhp-h for CO and 0.30 g/bhp-h for NMNEHC based on the weighted
average emission factors and the given control efficiencies.
Using the BAT emission rates, the assumed control efficiencies, and assuming full-year
operation, the control cost for SCR for lean-burn engines greater than or equal to 100 bhp but
less than 500 bhp is estimated between $25,843 and $62,940 per ton of NO
X
reduced. The
Department determines SCR is not BAT for lean-burn engines greater than or equal to 100 bhp
but less than 500 bhp because it is not economically feasible, and the emission limit remains
1.00 g/bhp-h for NO
X
as in the previously issued GP-5.
Lean-Burn Engines Greater Than or Equal To 500 bhp but Less Than 1,875 bhp
Vendor data for lean-burn engines greater than or equal to 500 bhp but less than 1,875 bhp
indicate uncontrolled weighted average emission rates of 1.4 g/bhp-h for NO
X
, 2.0 g/bhp-h for
CO, 3.8 g/bhp-h for THC, and 0.5 g/bhp-h for NMNEHC. Both the weighted average emission
rates and the current uncontrolled BAT for NO
X
were used in the BAT analysis.
Using the weighted average emission rates and BAT emission rates, the assumed control
efficiencies, and assuming full-year operation, the control cost for an oxidation catalyst for lean-
burn engines greater than or equal to 500 bhp but less than 1,875 bhp is estimated between
$232 and $2,081 per ton of pollutants reduced. Therefore, the Department determines oxidation
catalyst remains BAT for engines greater than 500 bhp but less than 1,875 bhp with an emission
limit of 0.25 g/bhp-h for CO based on the weighted average emission factor and the given
control efficiencies and an emission limit of 0.25 g/bhp-h for NMNEHC as in the previously
issued GP-5.
Using the BAT emission rates, the assumed control efficiencies, and assuming full-year
operation, the control cost for SCR for lean-burn engines greater than or equal to 500 bhp but
less than 1,875 bhp is estimated between $10,466 and $13,477 per ton of NO
X
reduced. SCR is
not considered as BAT for lean-burn engines greater than or equal to 500 bhp but less than
1,875 bhp because it is not economically feasible, and the emission limit remains 0.50 g/bhp-h
for NO
X
as in the previously issued GP-5.

02/04/2017
Page
29
of
73
Lean-Burn Engines Greater Than or Equal To 1,875 bhp
Vendor data for lean-burn engines greater than or equal to 1,875 bhp indicate uncontrolled
weighted average emission rates of 0.7 g/bhp-h for NO
X
, 2.1 g/bhp-h for CO, 5.7 g/bhp-h for
THC, and 0.8 g/bhp-h for NMNEHC.
Using the weighted average emission rates and BAT emission rates, the assumed control
efficiencies, and assuming full-year operation, the control cost for an oxidation catalyst for lean-
burn engines greater than or equal to 1,875 bhp is estimated between $123 and $873 per ton of
pollutants reduced. Therefore, the Department determines oxidation catalyst to be BAT for lean-
burn engines greater than 1,875 bhp with an emission limit of 0.25 g/bhp-h for CO based on the
weighted average emission factor and the given control efficiency and an emission limit of
0.25 g/bhp-h for NMNEHC as in the previously issued GP-5.
Using the current BAT emission rate of 0.50 g/bhp-h as baseline, the assumed control efficiency,
and assuming full-year operation, the control cost effectiveness for SCR for lean-burn engines
greater than 1,875 bhp is estimated to be less than $8,818 per ton of NO
X
reduced.
Department stack test data show that approximately 33% of engines greater than or equal to
1,875 bhp are capable of achieving a NO
X
emissions rate of 0.35 g/bhp-h uncontrolled. In
addition, recent conversations with engine vendors revealed that improvements to their engine
series that includes those rated at or above 1,875 bhp will have a vendor’s guarantee of 0.30
g/bhp-h. Therefore, the Department performed an additional analysis using 0.35 g/bhp-h of NO
X
as a baseline, the assumed control efficiency, and assuming full-year operation, and found the
control cost effectiveness for SCR for lean-burn engines greater than or equal to 1,875 bhp but
less than 3,000 bhp is estimated to increase to between $10,241 and $12,597. The control cost
for engines greater than or equal to 3,000 bhp, using a NO
X
emissions rate of 0.35 g/bhp-h as the
baseline, the assumed control efficiency, and assuming full-year operation, the control cost
effectiveness for SCR is estimated to be less than $9,064.
The Department therefore determines that lean-burn engines greater than or equal to 1,875 bhp
but less than 3,000 bhp have a dual BAT criterion of 0.35 g/bhp-h uncontrolled or 0.05 g/bhp-h
with control. The Department determines SCR to be BAT for engines greater than or equal to
3,000 bhp with a NO
X
emission limit of 0.05 g/bhp-h. Engines that utilize SCR to control NO
X
are required to limit ammonia slip to 5 ppmdv or less corrected at 15% O
2
.

02/04/2017
Page
30
of
73
The BAT emission limits for the proposed General Permits are summarized in the following
table:
Table 3 - BAT Emission Limits for Existing SI-RICE
Engine Type
Rated
bhp
NO
X
CO
NMNEHC as
propane
(excluding
HCHO)
HCHO
Permitted Under GP-5 Prior to Feb 2, 2013
NG-fired Lean- and
Rich-Burn Engines
<1,500
2.00 g/bhp-h
2.00 g/bhp-h
2.0 g/bhp-h
Permitted Under GP-5 On or After Feb 2, 2013 but Prior to (Insert Date)
NG-fired Lean- and
Rich-burn Engines
≤100
2.00 g/bhp-h
2.00 g/bhp-h
-
-
NG-fired Lean-burn
Engines
>100 to
≤500
1.00 g/bhp-h
2.00 g/bhp-h
0.70 g/bhp-h
-
NG-fired Lean-burn
Engines
>500
0.50 g/bhp-h
47 ppmvd @ 15%
O
2
or 93% reduction
0.25 g/bhp-h
0.05
g/bhp-h
NG-fired Rich-burn
Engines
>100 to
≤500
0.25 g/bhp-h
0.30 g/bhp-h
0.20 g/bhp-h
NG-fired Rich-burn
Engines
>500
0.20 g/bhp-h
0.30 g/bhp-h
0.20 g/bhp-h
2.7 ppmvd @ 15%
O
2
or 76% reduction
Table 4 - Proposed BAT Emission Limits for New SI-Rice
Engine Type
Rated
bhp
NO
X
CO
NMNEHC as
propane
(excluding
HCHO)
HCHO
Permitted On or After (Insert Date)
New NG-fired Lean-
burn Engines
≤100
1.00 g/bhp-h
2.00 g/bhp-h
0.20 g/bhp-h
-
New NG-fired Lean-
burn Engines
>100 to
≤500
1.00 g/bhp-h
0.70 g/bhp-h
0.30 g/bhp-h
-
New NG-fired Lean-
burn Engines
>500 to
<1,875
0.50 g/bhp-h
0.25 g/bhp-h
0.25 g/bhp-h
0.05 g/bhp-h
New NG-fired Lean-
burn Engines
≤1,875
to
<3,000
0.35 g/bhp-h
Uncontrolled
or 0.05 g/bhp-h
with Control
0.25 g/bhp-h
0.25 g/bhp-h
0.05 g/bhp-h
New NG-fired Lean-
burn Engines
≥3,000
0.05 g/bhp-h
0.25 g/bhp-h
0.25 g/bhp-h
0.05 g/bhp-h
New NG-fired Rich-
burn Engines
≤100
0.60 g/bhp-h
0.60 g/bhp-h
0.10 g/bhp-h
-
New NG-fired Rich-
burn Engines
>100 to
≤500
0.25 g/bhp-h
0.30 g/bhp-h
0.20 g/bhp-h
New NG-fired Rich-
burn Engines
>500
0.20 g/bhp-h
0.30 g/bhp-h
0.20 g/bhp-h
2.7 ppmvd @
15% O
2
or 76%
reduction

02/04/2017
Page
31
of
73
In addition, the engines shall comply with all applicable requirements specified in 40 CFR Part
60, Subpart JJJJ and 40 CFR Part 63, Subpart ZZZZ.
For engines constructed on or after June 12, 2006, compliance with the requirements in the
following table guarantees compliance with the requirements of 40 CFR Part 60 Subpart JJJJ and
40 CFR Part 63 Subpart ZZZZ:
Table 5 - 40 CFR Part 60 Subpart JJJJ Requirements
Engine Size
Manufacture
Date
NO
X
NMNEHC
(as propane)
excluding
HCHO
CO
≤25 bhp and <225 cc
displacement
7/1/2008
12.01 g/bhp-h
387.02 g/bhp-h
1/1/2012
7.46 g/bhp-h
454.88 g/bhp-h
≤25 bhp and ≥225 cc
displacement
7/1/2008
9.99 g/bhp-h
387.02 g/bhp-h
1/1/2011
5.97 g/bhp-h
454.88 g/bhp-h
25 bhp < ER < 100 bhp
1/1/2007
2.83 g/bhp-h
4.85 g/bhp-h
1/1/2011
1.00 g/bhp-h
0.70 g/bhp-h
2.00 g/bhp-h
100 hp ≤ ER < 500 bhp
7/1/2008
2.00 g/bhp-h
1.00 g/bhp-h
4.00 g/bhp-h
1/1/2011
1.00 g/bhp-h
0.70 g/bhp-h
2.00 g/bhp-h
≥500 bhp
7/1/2007
2.00 g/bhp-h
1.00 g/bhp-h
4.00 g/bhp-h
7/1/2010
1.00 g/bhp-h
0.70 g/bhp-h
2.00 g/bhp0h
However, for all previous versions of the GP-5, the Department’s BAT requirements are more
stringent than those listed in the table above. Therefore, by complying with the Department’s
BAT requirements, the owner or operator of an engine will be guaranteed compliance with the
applicable emission limits of 40 CFR Part 60 Subpart JJJJ and 40 CFR Part 63 Subpart ZZZZ for
new engines.

02/04/2017
Page
32
of
73
For engines constructed prior to June 12, 2006, compliance with the requirements in the
following table guarantees compliance with the requirements of 40 CFR Part 63 Subpart ZZZZ:
Table 6 - 40 CFR Part 63 Subpart ZZZZ Requirements
Engine Category
Oil and
Filter
Change
Spark Plug
Inspection,
Plugs Replaced as
Necessary
Hose Inspection,
Hoses Replaced as
Necessary
Emergency SI RICE; 4SRB
and 4SLB >500 hp that
operate ≤ 24 hours per year
500 hours or
annually
1,000 hours or annually
500 hours or annually
4SRB and 4SLB > 500 hp
in remote locations
2,160 hours or
annually
2,160 hours or annually
2,160 hours or annually
4SLB > 500 hp
Install Oxidation Catalyst to Reduce HAP Emissions
4SRB > 500 hp
Install NSCR to Reduce HAP Emissions
4SRB and 4SLB ≤ 500 hp
1,440 hours or
annually
1,440 hours or annually
1,440 hours or annually
2SLB
4,320 hours or
annually
4,320 hours or annually
4,320 hours or annually
Visible emissions shall not meet or exceed 10% opacity for a period or periods aggregating more
than three minutes in any one hour nor meet or exceed 30% opacity at any time.
Reciprocating Natural Gas Compressors
Fluids, such as natural gas, travel naturally from areas of high pressure to areas of low pressure.
Natural gas compressors take advantage of this property by increasing the pressure of natural gas
at one location in a pipeline in order to promote the movement of the gas to a lower pressure area
downstream. The pipeline pressure tends to drop over the length of a pipeline due to friction.
This decrease in pressure is the reason why compressor stations are located along the length of
the pipeline.
Reciprocating natural gas compressors provide this increase of pressure by using a piston and
cylinder arrangement. As the piston moves through the chamber, the pressure at the forward
edge of the piston is increased as the volume in the cylinder is decreased. The high pressure gas
is then forced through a valve into the high pressure section of the pipeline. On the reverse
stroke, the pressure at the trailing edge of the piston is decreased as the volume in the cylinder is
increased. This reduction in pressure allows the low pressure gas in the pipeline to be drawn into
the cylinder. The piston is connected to its prime mover by a rod, and the rod utilizes rod
packings to reduce wear on the compressor components and to seal in the gas pressure. Over
time, these packings can wear, resulting in methane, VOC, and HAP emissions.
It is not typical for reciprocating compressors to be installed at an unconventional natural gas
well site. According to 40 CFR Part 60, Subparts OOOO and OOOOa, reciprocating
compressors located at well sites are not affected facilities. The Department is proposing to
require reciprocating compressors located at well sites to meet the same requirements for
reciprocating compressors located at compressor stations as BAT.

02/04/2017
Page
33
of
73
Existing Reciprocating Natural Gas Compressors
40 CFR Part 60, Subpart OOOO provides two options for controlling VOC emissions from
reciprocating natural gas compressors. The first is to replace the rod packings either every
26,000 hours of operation (operating hours must be monitored and documented) or every
36 months (monitoring and documentation of operating hours not required). The second is to
utilize a rod packing emissions collection system that operates under negative pressure to route
the rod packing emissions to a process through a closed vent system.
The Department previously determined the requirements of 40 CFR Part 60, Subpart OOOO to
be BAT. Therefore, the owner or operator of an existing reciprocating natural gas compressor
shall continue to comply with the applicable requirements specified in 40 CFR Part 60,
Subpart OOOO, which are detailed in the General Permit.
New Reciprocating Natural Gas Compressors
40 CFR Part 60, Subpart OOOOa has identical requirements to 40 CFR Part 60, Subpart OOOO
for reciprocating natural gas compressors. Both control strategies mentioned above affect
methane and VOC; therefore, new requirements were not introduced.
The Department determines that the requirements of 40 CFR Part 60, Subpart OOOOa to be
BAT. Therefore, the owner or operator of a new reciprocating natural gas compressor shall
comply with the applicable requirements specified in 40 CFR Part 60, Subpart OOOOa, which
are detailed in the General Permit.
Glycol Dehydration Units and Associated Equipment
All natural gas well streams contain water vapor as they leave the reservoir, and this water is
produced along with the natural gas. As the natural gas travels up the well bore, it cools as a
result of pressure reduction and the conduction of heat through the casing to cooler formations.
Therefore, since the ability of gas to hold water vapor decreases as the gas temperature
decreases, natural gas is nearly always saturated with water vapor when it reaches surface
equipment. Additional cooling of the saturated gas will cause the formation of free water. The
process for removal of water vapor from natural gas is known as dehydration.
Dehydrators are designed to remove water from the natural gas vapor stream, thereby reducing
corrosion and preventing the formation of hydrates, which are solid compounds that can cause
flow restrictions and plugging in valves and even pipelines. The water-lean glycol usually flows
downward in an absorption tower, counter-current to the natural gas. The glycol absorbs most of
the water from the natural gas, but it also absorbs other materials present in the gas stream. The
dried natural gas exits the top of the tower. The water-rich glycol leaves the bottom of the tower
and flows to the regenerator. The regenerator heats the glycol to drive off water vapor, and the
water vapor is usually vented directly to the atmosphere through the regenerator vent stack.
While water has a boiling point of 212 degrees Fahrenheit, glycol does not boil until 400 degrees
Fahrenheit. This difference in the boiling points allows for the easy removal of water from the
glycol. The water-lean glycol is then returned to the absorber. Glycol has a high affinity for
water and a relatively low affinity for non-aromatic hydrocarbons, which makes it a very good
absorbent fluid for drying natural gas. However, the glycol does absorb small amounts of
methane and other hydrocarbons from the natural gas. The hydrocarbons are released to the
atmosphere along with the water vapor from the regenerator vent.

02/04/2017
Page
34
of
73
Some glycol dehydrators have additional equipment. Two common additions are flash tanks and
regenerator vent emissions control equipment. The flash tank is placed in the rich glycol loop
between the absorber and the regenerator. The glycol line pressure is dropped in the flash tank,
causing most of the light hydrocarbons to flash into the vapor phase. The flash gas is usually
routed to the regenerator burner as fuel. The methane emissions from the regenerator vent can
be significantly reduced by using a flash tank. Regenerator vent control devices on units reduce
emissions of BTEX and VOC to the atmosphere. These compounds are absorbed from the gas
stream and driven off with the water in the regenerator vent. Control devices usually condense
the water and hydrocarbons (containing BTEX and heavier VOC), then decant the hydrocarbons
for sale and the water for disposal.
Emissions from glycol dehydration units are often controlled by using a condenser on the
regenerator still vent and then venting to the atmosphere or to the regenerator firebox, other
heaters, or a flare. Emissions from water-rich glycol flash tank vents are often controlled by
combustion or by recycling back to low-pressure inlet gas streams. According to the Department
of Energy's Office of Fossil Energy, these systems have been shown to recover 90 to 99 percent
of methane that would otherwise be flared into the atmosphere.
7
Emission Limits for Glycol Dehydrators
Existing Glycol Dehydrators
The owner or operator of each existing glycol dehydrator located at an unconventional natural
gas well site, remote pigging station, natural gas compressor station, natural gas processing plant,
or natural gas transmission station shall comply with the applicable requirements established in
40 CFR Part 63, Subpart HH, which are detailed in the General Permit. It is important to note
that the locations listed in the technical support documents of 40 CFR Part 63, Subpart HH as
urbanized areas and urban clusters were updated in the 2010 Census. The new list can be found
in Appendix F.
The owner or operator of each existing glycol dehydrator installed prior to February 2, 2013,
which has a total uncontrolled PTE of VOC in excess of 10 tpy, shall be controlled by at least
85% with a condenser, enclosed flare, or other air cleaning device approved by the Department.
The owner or operator of each existing glycol dehydrator installed on or after February 2, 2013,
but before (
insert date
) at a natural gas compressor station, processing plant, or transmission
station, which has a total uncontrolled PTE of VOC in excess of 5 tpy, shall be controlled by at
least 95% with a condenser, enclosed flare or other air cleaning device approved by the
Department.
The owner or operator of each existing glycol dehydrator installed on or after August 10, 2013,
but before (
insert date
) at an unconventional natural gas well site or remote pigging station,
which has a total uncontrolled VOC emission rate greater than or equal to 2.7 tpy, an
uncontrolled single HAP emission rate greater than or equal to 0.5 tpy, or a total HAP emission
rate greater than or equal to 1.0 tpy, shall be controlled by at least 95% with a condenser,
enclosed flare, or other air cleaning device approved by the Department.
7
Web Site:
http://naturalgas.org/naturalgas/processing-ng/#water
.

02/04/2017
Page
35
of
73
New Glycol Dehydrators
The owner or operator of each new glycol dehydrator located at an unconventional natural gas
well site, remote pigging station, natural gas compressor station, processing plant, or
transmission station shall comply with the applicable requirements established in 40 CFR
Part 63, Subpart HH, which are detailed in the General Permit.
For new glycol dehydrators, whose uncontrolled emissions are below the methane
de minimis
level of 200 tpy, VOC
de minimis
level of 2.7 tpy, single HAP limit of 0.5 tpy, and combined
HAP limit of 1 tpy, no controls are required.
For new glycol dehydrators, whose uncontrolled emissions are greater than or equal to the
methane
de minimis
level of 200 tpy, VOC
de minimis
level of 2.7 tpy, single HAP limit of
0.5 tpy, or combined HAP limit of 1 tpy, emission controls must be installed and emissions must
be reduced by at least 98%. A cost analysis was done by the Department and is included in
Appendix D. The control requirements are summarized in the table below.
Table 7 - Glycol Dehydrator Control Thresholds and Control Requirements
Uncontrolled
VOC PTE
Still Vent Control
Level
Permitted Under GP-5 Prior to Feb 2, 2013
>10 tpy
85%
Permitted Under GP-5 On or After Feb 2,
2013 but Prior to (Insert Date)
>5 tpy
95%
Permitted On or After (Insert Date)
≥2.7 tpy
98%
Storage Vessels
Storage vessels are used to collect and store condensate (NGLs) and/or produced water that are
byproducts of natural gas production. Most storage vessels in the natural gas industry are fixed
roof structures, and are equipped with a variety of pressure equalization devices to protect the
structural integrity of the tank.
There are several federal regulations that pertain to storage vessels including requirements found
in 40 CFR Part 60, Subparts K, Kb, OOOO, and OOOOa and 40 CFR Part 63, Subpart HH. In
addition, state regulations found in 25 Pa. Code §§ 129.56 and 129.57 have applicable
requirements. 40 CFR Part 60, Subparts OOOO and OOOOa include inspection and monitoring
requirements for storage vessels; this includes the requirement that the owner or operator must
conduct the no detectable emissions test procedure in accordance with 40 CFR Part 60,
Appendix A-7, Method 21.

02/04/2017
Page
36
of
73
Emission Limits for Storage Vessels
Existing Storage Vessels
Based on the conditions in Category Number 38 of the Air Quality Permit Exemption document,
storage vessels at unconventional natural gas well sites shall be equipped with controls achieving
VOC emissions reductions of 95% or greater unless the facility’s uncontrolled VOC emissions
for all sources are below 2.7 tpy. If the storage vessels’ HAP emissions are uncontrolled, they
must also be included in the facility-wide uncontrolled single HAP emissions limit of 0.5 tpy and
total uncontrolled HAP emissions limit of 1.0 tpy. The facility-wide uncontrolled emission
limits do not include emissions from any source that is equipped with emission controls.
Existing storage vessels built on or after July 23, 1984, but prior to August 23, 2011, are subject
to 40 CFR Part 60 Subpart Kb, which required storage vessels above 19,812 gallons (75 m
3
) to
either install a fixed roof tank with internal floating roof, an external floating roof, or a fixed roof
tank with a closed vent system routed to a control device that achieved 95% or greater control.
These conditions were not included in the General Permit, however, as the majority of storage
vessels located at unconventional natural gas well sites, remote pigging stations, natural gas
compressor stations, processing plants, and transmission stations are either smaller than the listed
capacity, or were built on or after August 23, 2011.
Existing storage vessels built on or after August 23, 2011, but prior to (insert date), with
uncontrolled potential VOC emissions greater than or equal to 6.0 tpy are required to be
controlled by 95% or more, install a fixed roof tank with an internal floating roof, an external
floating roof, or maintain the actual uncontrolled VOC emissions below 4.0 tpy. In addition, any
storage vessel greater than 2,000 gallons and less than or equal to 40,000 gallons are required to
install pressure relief valves in accordance with 25 Pa. Code §129.57, which are detailed in the
General Permit.
New Storage Vessels
Similar to the conditions in Category Number 38 of the Air Quality Permit Exemption document,
storage vessels shall be equipped with controls achieving methane, VOC, and HAP emissions
reductions of 98% or greater unless their uncontrolled methane, total VOC, single HAP, and total
HAP emissions are below 200 tpy, 2.7 tpy, 0.5 tpy, and 1.0 tpy, respectively. See the cost
analysis in Appendix D. In addition, the owner or operator of any storage vessel greater than
2,000 gallons and less than or equal to 40,000 gallons is required to install pressure relief valves
in accordance with 25 Pa. Code §129.57, which is detailed in the General Permit.
Tanker Truck Load-Out Operations
The storage tanks at unconventional natural gas well sites, natural gas compression facilities, and
natural gas processing facilities must be unloaded on occasion. This is done by loading the
liquids from the tanks into tanker trucks so the liquids may be transported to a processing
facility. The unloading process may emit methane, VOC, and HAP.
All tanker truck load-out operations are required to use a vapor recovery load-out system that
meets the closed vent system requirements in the section on enclosed flares and other control
devices.

02/04/2017
Page
37
of
73
When calculating the emissions from tanker truck load-out operations, the collection efficiency
may be assumed to be 99.2% for tanker trucks that pass the MACT-level annual leak test and
98.7% for tanker trucks that pass the NSPS-level annual test. The MACT-level leak test is
passed if the tanker does not indicate more than a 1” H
2
O pressure change within 5 minutes after
being pressurized to 18” H
2
O and after being depressurized to 6” H
2
O vacuum. The NSPS-level
leak test is passed if the tanker does not indicate more than a 3” H
2
O pressure change within
5 minutes after being pressurized to 18” H
2
O and after being depressurized to 6” H
2
O vacuum.
A leak test performed in accordance with 49 CFR §180.407 –
Requirements for Test and
Inspection of Specification Cargo Tanks
, or EPA Method 27 –
Determination of Vapor Tightness
of Gasoline Delivery Tank Using Pressure Vacuum Test
, will be accepted as equivalent to an
NSPS-level collection efficiency (i.e., 98.7%).
To facilitate the emissions calculations, the owner or operator is required to keep records of the
entire fleet of tanker trucks that collect liquids from the facility, including the date and rating of
each leak test and an identification number for each truck. The fleet records are then cross
referenced with the load-out records, which identify the truck performing the load-out, the date
and time the load-out occurred, and the type and volume of liquids loaded. These records can
then be used to calculate emissions due to the load-out operations for the emissions inventory.
Fugitive Emissions Components
Equipment leaks are typically low-level, unintentional losses of process gas from the sealed
surfaces of above-ground process equipment. Equipment components that tend to leak include
valves, flanges and other connectors, pump seals, compressor seals, pressure relief valves, open-
ended lines, and sampling connections. These components represent mechanical joints, seals,
and rotating surfaces, which in time tend to wear and develop leaks. However, a release from
any equipment or component designed by the manufacturer to protect the equipment, controller,
or personnel or to prevent groundwater contamination, gas migration, or an emergency situation
is not considered a leak. The following requirements have been included to minimize and/or
eliminate the equipment leaks.
In accordance with the Department’s requirements under the previous version of GP-5, the
owner or operator of a natural gas compression facility and/or natural gas processing facility
shall, at a minimum on a monthly basis, perform a leak detection and repair program which
includes auditory, visual, and olfactory (AVO) inspections. This requirement is proposed to be
extended to unconventional natural gas well sites, remote pigging stations, and natural gas
transmission stations covered by the General Permits.
In the previous version of the GP-5, the owner or operator of the facility was required to use an
optical gas imaging (OGI) camera or other leak detection device to conduct a leak detection and
repair (LDAR) program inspection within 180 days after the initial startup of a source and at a
minimum of once a quarter thereafter. This requirement has been changed to within 60 days
after the initial startup of a source to meet the 40 CFR Part 60 Subpart OOOOa requirement.
The LDAR inspection for unconventional natural gas well sites and remote pigging stations is
required to be performed within 60 days of the start of production, and the Department proposes
subsequent inspections be performed at a minimum of once a quarter thereafter. An owner or
operator of an unconventional natural gas well site or remote pigging station may track the
number of leaking components in the LDAR program and reduce the inspection interval from
once per quarter to semi-annually if the percentage of leaking components is less than 2.0% for

02/04/2017
Page
38
of
73
two consecutive inspections. If the percentage of leaking components is higher than 2.0% in any
inspection, the quarterly LDAR inspection interval must be resumed or maintained.
Consistent with 40 CFR Part 60, Subparts KKK, OOOO, and OOOOa, the LDAR requirements
for natural gas processing plants are to use Method 21. As per 40 CFR §65.7(e), when a
Method 21 inspection is required in any subpart of Parts 60, 61, 63, and 65, OGI camera
inspections are an accepted alternative work practice for monitoring equipment for leaks.
Therefore, the Department proposes that all LDAR inspections covered by GP-5 and GP-5A can
use the same criteria, including those at natural gas processing plants under 40 CFR §§60.484
and 60.484a.
Also, consistent with 40 CFR Part 60 Subpart OOOOa, an existing facility that undergoes a
modification, i.e. through the drilling and/or hydraulic fracturing or refracturing of a well or the
addition of a new piece of equipment must use the new LDAR requirements of the General
Permit.
A leak is defined as any positive indication, whether audible, visual, or odorous, determined
during an AVO inspection, any visible emission detected by an OGI camera, or a concentration
of 500 ppm or greater detected by an instrument reading, regardless of source.
If any leak is detected, the owner or operator of the facility shall make a first attempt of repair
within five days of the detection of the leak. The leak must be repaired no later than 15 days
after the leak is detected, unless the repair requires the ordering of parts, in which case the repair
must be completed no later than 10 days after the receipt of the parts, or if the repair is
technically infeasible without a vent blowdown, facility shutdown, or well shut-in or would be
unsafe to repair during operation of the unit, in which case the repair must be completed at the
earliest of the next scheduled or unscheduled blowdown, or within two years.
A leak is considered repaired if one of the following can be demonstrated:
?
No detectable emissions consistent with 40 CFR Part 60, Appendix A-7,
Method 21 Section 8.3.2;
?
A concentration of less than 500 ppm is detected when the gas leak detector probe
inlet is placed at the surface of the component;
?
No visible leak image when using an OGI camera;
?
No bubbling at leak interface using a soap solution bubble test specified in
Section 8.3.3 of 40 CFR Part 60, Appendix A-7, Method 21.
LDAR is considered to have a fugitive emission control rate based on the frequency of the
inspection. According to EPA and Colorado
8
the emissions reductions from annual LDAR is
40% and from quarterly LDAR is 60%. In 40 CFR Part 60, Subpart OOOOa, the LDAR is
required semi-annually for well pads and quarterly for compressor stations and processing plants.
The Department assumes that the semi-annual requirement will yield 50% emissions reductions.
These emissions reduction assumptions are used in the LDAR cost analysis which is in
Appendix E.
8
ICF International, Economic Analysis of Methane Emission Reduction Opportunities in the U.S. Onshore Oil and
Natural Gas Industries, March 2014.

02/04/2017
Page
39
of
73
Controllers
Controllers are automated instruments used for maintaining liquid levels, pressure, and
temperature at unconventional natural gas well sites, remote pigging stations, natural gas
compression stations, natural gas processing plants, and natural gas transmission stations. These
controllers often are powered by high-pressure natural gas and may release methane, VOC, and
HAP with every valve movement (i.e., intermittent bleed), or continuously (i.e., continuous
bleed) as part of their normal operations.
Existing controllers shall comply with the applicable requirements specified in 40 CFR Part 60,
Subpart OOOO. For pneumatic controllers located at unconventional natural gas well sites,
remote pigging stations, natural gas compressor stations, and natural gas transmission stations
this means they must be low-bleed controllers with an emission rate less than or equal to
6.0 standard cubic feet per hour unless a higher bleed rate is required for operational reasons
such as speed, safety, or positive actuation.
New controllers constructed on or after (insert date) shall either be an electric controller if the
facility has access to electricity on site or meet the requirements of 40 CFR Part 60
Subpart OOOOa if electricity is not available on site. For pneumatic controllers complying with
Subpart OOOOa located at unconventional natural gas well sites, remote pigging stations, natural
gas compressor stations, and natural gas transmission stations this means they must be low-bleed
controllers with an emission rate less than or equal to 6.0 standard cubic feet per hour unless a
higher bleed rate is required for operational reasons as above.
The owner or operator of new and existing controllers located at a natural gas processing plant
shall employ no-bleed pneumatic controllers. These can be electrically actuated controllers or
pneumatic controllers driven by instrument air. Natural gas actuated controllers that route the
emissions into the downstream pipeline can also be used.
Pumps
Pumps are primarily used at unconventional natural gas well sites, remote pigging stations,
natural gas compressor stations, processing plants, and transmission stations for glycol
circulation or for injecting chemicals used in normal operations. Pneumatic pumps use
pressurized air or natural gas to operate the pump; at natural gas facilities, it is common to use
natural gas from the production stream to operate the pumps. The pressurized natural gas, after
being used to operate the pump, is often vented to the atmosphere through the exhaust port.
There are many options available to reduce or eliminate emissions to the atmosphere.
Pneumatic pumps had no standards in 40 CFR Part 60, Subpart OOOO other than the LDAR
requirements for pumps at natural gas processing facilities. In 40 CFR Part 60, Subpart OOOOa,
natural gas-driven diaphragm pumps have control requirements for GHG and VOC depending
upon the type of facility at which they are located and the number of days they are operated.
Subpart OOOOa does not have a requirement for pumps located at natural gas compressor
stations.
The Department requires that electric pumps be used at any facility other than a natural gas
processing plant that has access to electricity on site. For facilities that do not have access to
electricity on site, the Department proposes that the requirements of 40 CFR Part 60,
Subpart OOOOa, which are detailed in the General Permit, are BAT for pumps located at

02/04/2017
Page
40
of
73
unconventional natural gas well sites and natural gas processing plants. The Department also
proposes that the requirements for pumps at well sites are also BAT for pumps located at remote
pigging stations, natural gas compressor stations, and transmission stations.
The Department is also proposing to collect information on other types of pneumatic pumps by
having notification, recordkeeping, and reporting requirements for all pneumatic pumps.
Enclosed Flares and Other Control Devices
Most of the BAT requirements for emissions sources are dependent on a control to reduce those
emissions to the atmosphere. The conditions for operating, maintaining, and performance testing
those control devices are included in this section of the General Permit. The majority of the
requirements listed are from EPA rules, most notably 40 CFR Part 60 Subpart OOOOa.
In the EPA rules, the control requirement is generally 95% or better for methane and/or VOC.
The Department is proposing, however, that a control requirement of 98% is achievable and
reasonable based on the economic feasibility of combustion control devices, as shown in
Appendix D, and the demonstration below that at a combustion zone temperature of 1,600 °F a
methane destruction of 98% is achievable.
For the demonstration of combustion zone temperature to assure greater than 98% destruction
efficiency, the following equations from
Air Pollution Control: A Design Approach
are used.
9
㴋㴋㴋
䰍 眃笃瘃 䘍 猃球球㤇
䔍 猃猃礃爃㤇
䔍 礃猃砃㤇
䔍 稃爃球㤇
䔍 爃眃笃球㤇
䘍 球爃球㤇
䘍 瘃球爃甃㤇
䔍 稃礃猃㤇
䘍 砃砃稃㤇
䔍 砃球稃㤇
㔋㐋
䘍 礃眃甃㤇
㔋㔋
㴋㴋
䰍 眃礃礃 䘍 猃爃爃㤇
䔍 猃猃爃球㤇
䔍 砃礃猃㤇
䔍 礃球砃㤇
䔍 爃眃稃砃㤇
䘍 球甃瘃㤇
䘍 瘃甃爃笃㤇
䔍 稃眃球㤇
䘍 稃球球㤇
䔍 砃眃眃㤇
㔋㐋
䘍 礃砃猃㤇
㔋㔋
Where:
T
99.9
= Temperature for 99.9% destruction efficiency, °F
T
99
= Temperature for 99% destruction efficiency, °F
W
1
= Number of Carbon Atoms
W
2
= Aromatic Compound Flag (no = 0, yes = 1)
W
3
= Carbon Double Bond Flag [Exclude Aromatic Ring] (no = 0, yes = 1)
W
4
= Number of Nitrogen Atoms
W
5
= Autoignition Temperature, °F
W
6
= Number of Oxygen Atoms
W
7
= Number of Sulfur Atoms
W
8
= Hydrogen to Carbon Ratio
W
9
= Allyl (2-Propenyl) Compound Flag (no = 0, yes = 1)
W
10
= Carbon Double Bond – Chlorine Interaction Flag (no = 0, yes = 1)
W
11
= Natural Logarithm of Residence Time, s
Using these equations for propane (C
3
H
8
), which has an autoignition temperature of 871 °F, and
assuming a residence time of 0.3 s, we find:
9
Air Pollution Control: A Design Approach, Fourth Edition, Cooper, C. David, et. al. Waveland Press, Inc., 2011.

02/04/2017
Page
41
of
73
㴋㴋㴋
䰍 眃笃瘃 䘍 猃球球㨒甃㬒 䔍 爃眃笃球㨒稃礃猃㬒 䔍 稃礃猃 氍
瀍 䘍 礃眃甃㨒踀退㨒爃甃㬒㬒 䰍 猃甃笃砃̀뤃⠇
㴋㴋
䰍 眃礃礃 䘍 猃爃爃㨒甃㬒 䔍 爃眃稃砃㨒稃礃猃㬒 䔍 稃眃球 氍
瀍 䘍 礃砃猃㨒踀退㨒爃甃㬒㬒 䰍 猃甃礃砃̀뤃⠇
This matches EPA’s temperature requirements of 1,400 °F for combustion zone temperature for
VOCs.
Using these equations for methane (CH4), which has an autoignition temperature of 999 °F, and
assuming a residence time of 0.3 s, we find:
㴋㴋㴋
䰍 眃笃瘃 䘍 猃球球㨒猃㬒 䔍 爃眃笃球㨒笃笃笃㬒 䔍 稃礃猃 氍
瀍 䘍 礃眃甃㨒踀退㨒爃甃㬒㬒 䰍 猃砃猃球̀뤃⠇
㴋㴋
䰍 眃礃礃 䘍 猃爃爃㨒猃㬒 䔍 爃眃稃砃㨒笃笃笃㬒 䔍 稃眃球 氍
瀍 䘍 礃砃猃㨒踀退㨒爃甃㬒㬒 䰍 猃眃稃瘃̀뤃⠇
Assuming a linear relationship between temperature and destruction efficiency, this implies that
at the 1,400 °F combustion zone temperature required in Subpart OOOOa, which assures VOC
(as propane) destruction of up to 99.9%, the methane destruction rate is approximately 93%.
Methane would require a combustion zone temperature of 1,552 °F to ensure 98% destruction
efficiency. Therefore, to ensure that the 98% destruction efficiency is met or exceeded for
methane, the Department is altering the requirement that the minimum combustion zone
temperature of 1,400 °F required by the EPA for enclosed combustion devices be raised to
1,600 °F.
Pigging Operations
Pigging operations are undertaken to remove accumulated water and condensate liquids in
natural gas gathering pipelines. This operation is done as necessary to maintain optimal pressure
in the pipeline that keeps the natural gas flowing and to push valuable condensate to tanks where
it can be transported to a processing plant. The “pig” is a spherical or bullet shaped device that
travels through the pipeline to push the liquids to their eventual destination
A pig must be loaded into the pipeline at a launching station and recovered at a receiving station.
When the pig is launched and recovered, some of the natural gas in the chamber is vented to the
atmosphere. This venting can be reduced by routing the gas to a vapor recovery system or a
flare. It can also be minimized in high-pressure pipelines by equalizing the high-pressure
chamber with a low-pressure line before venting. The EPA’s Partner Reported Opportunities
(PRO) for Reducing Methane Emissions, also called the Natural Gas Star program or NGStar,
PRO Fact Sheet No. 505, gives information on ways to minimize emissions from pigging
operations.

02/04/2017
Page
42
of
73
Figure 1: Drawing of a barrel type design of a pig launcher and receiver equipped with uncontrolled
depressurization vents.
10
The Department requires that the owner or operator of each pigging-affected facility calculate
the total annual emissions of VOC and methane using the worksheet located at:
http://files.dep.state.pa.us/Air/AirQuality/AQPortalFiles/Business%20Topics/Emission%20Inven
tory/marcellus/Midstream%20Pigging%20Spreadsheet.xlsx
.
The owner or operator of a pigging operation shall minimize all emissions to the atmosphere to
the highest extent possible. All pig receiver chambers must be equipped with a liquids drain.
All high-pressure pig launcher and receiver chambers must be vented to a low-pressure pipeline
or vessel if available. An owner or operator of a pigging operation whose uncontrolled
emissions exceed the methane
de minimis
level of 200 tpy, the total VOC de minimis level of
2.7 tpy, the single HAP de minimis level of 0.5 tpy, or the combined HAP de minimis level of
1.0 tpy shall control emissions by 98% or more.
Sources Specific to GP-5A:
Well Drilling and Hydraulic Fracturing Operations
The first step of establishing a natural gas well is the drilling phase. This typically entails
drilling the wellbore in stages using either a diesel engine rig or a natural gas engine rig.
Because the drill rig is on site for a short period of time, the only standards that apply are that
any drill rig engine must meet the applicable non-road engine standards. However, the owner or
operator must provide at least 24 hours advance notification to the Air Program Manager of the
appropriate Regional Office before drilling. After the wellbore is completed and encased, the
wellhead or “Christmas Tree” is installed.
After the installation of the wellhead, the well is ready for hydraulic fracturing. Large volumes
of water, mixed with chemicals and proppants are pumped into the formation to create and hold
open fractures in shale. This fracturing technique greatly enhances natural gas production.
Again, the owner or operator must provide at least 24 hours advance notification to the Air
10
Quantifying the Potential Impact of Natural Gas Condensate Holdup on Uncontrolled Volatile Organic
Compound Emissions from Pig Receivers During Depressurization in Wet Gas Gathering Operations
, EPA
Discussion Draft, May 2016.

02/04/2017
Page
43
of
73
Program Manager of the appropriate Regional Office before fracture. The trucks responsible for
mixing the fracturing fluid and pumping underground also use non-road engines and must meet
the applicable non-road engine standards. These are summarized in the tables in Appendix G.
Well Completion Operations
After a natural gas well is hydraulically fractured, the well has to be prepared for production of
natural gas. The fluid used in the fracturing process has to be removed from the well. During
this process, equipment such as a separator is used to separate and remove the sand and water
from the natural gas stream. The separated gas, instead of being vented, is either captured and
routed to a sales pipeline, or is flared. By not directly venting the gas to the atmosphere, both
methane and VOC emissions are greatly reduced. This method of completing a natural gas well
is called reduced emission completion (REC) or green completion.
The owner or operator shall use reduced emission completion methods in accordance with
requirements specified in 40 CFR Part 60, Subpart OOOOa. Also, any existing well that is
refractured after the applicability date of Subpart OOOOa subjects existing sources at a facility
to the fugitive emissions components requirements in this general permit.
Wellbore Liquid Unloading
Over time, liquids may accumulate in a producing natural gas well and may reduce the well
pressure to the point where production is reduced, especially in wells located in the wet gas
areas. When this happens, the accumulated fluids need to be removed in order to restore
production through a process called liquids unloading. There are many techniques that can be
used to accomplish this, including venting, soaping, swabbing, and using a plunger lift system.
As indicated in EPA’s white paper on Oil and Natural Gas Sector Liquids Unloading Processes,
published April, 2014, the use of technologies like a plunger lift system can reduce the frequency
of liquids unloading operations.
However, the basic criteria for the installation of a plunger lift, as found in the EPA’s NGStar
program are as follows:
?
Wells must produce at least 400 scf of gas per barrel of fluid per 1,000 feet of depth.
?
Wells with shut-in wellhead pressure that is 1.5 times the sales line pressure.
?
Wells with scale or paraffin buildup.
These conditions may not be found at all wells covered by the General Permit, so plunger lift
systems remain an option, but not a requirement for wellbore liquids unloading. Furthermore, as
indicated in API’s comments to EPA about why plunger lifts should not be required, they said:
It is a misconception that plunger-lift systems are the single emission control action for
wells where venting for liquids unloading occurs.
This misconception is exacerbated by a
lack of understanding, even among those purporting plunger lift systems as
the
solution to
liquids unloading, of liquids loading or plunger lift systems and their appropriate uses,
limitations, and efficacy. Plungers work by providing a mechanical barrier between a small
volume of water and the gas that is used to transport it up the well-bore. The mechanical barrier
isolates the gas from the liquids, prevents gas from moving up through the liquids hence making

02/04/2017
Page
44
of
73
better use of the gas energy, and helps prevent liquids from falling back into the well-bore. If the
gas could flow faster, then that mechanical barrier would not be necessary or helpful.”
“Although plungers are among the most common tools used in middle stage deliquification, there
is a misconception that plungers eliminate the need to vent to atmosphere. In many cases, wells
are vented to atmosphere to generate the differential pressure necessary to lift the plunger and
liquid column up the well-bore. While this can be controlled and minimized, it cannot be
eliminated.”
“As the API/ANGA report and the GHGRP data show, venting of wells to aid liquids unloading
occurs in both plunger equipped wells and non-plunger equipped wells with plunger equipped
wells having higher reported emissions overall.”
In conversations with some industry representatives, it was suggested that one of the largest
factors in emissions from wellbore liquids unloading events is the length of time that flow is
directed to atmospheric pressure. It was further suggested that one of the most effective methods
to reduce the time venting to atmospheric pressure would be to ensure that an operator remains
on site for the duration of a manual unloading operation. The Department concurs that this work
practice is an effective and reasonable requirement.
Therefore, the Department is proposing that the owner or operator shall ensure that an operator
remains on site for the duration of a manual unloading operation and that the owner or operator
use BMP including, but not limited to, a plunger lift system, soaping, swabbing, or venting to
atmospheric pressure to minimize methane and VOC emissions during wellbore liquids
unloading operations. In all cases, where technically feasible, the owner or operator shall direct
the gas to a separator, storage vessel, or control device.
Sources Specific to GP-5:
Natural Gas-Fired Simple Cycle Turbines
A simple cycle turbine is an internal combustion engine that operates with rotary rather than
reciprocating motion. A turbine is composed of three major components: the compressor, the
combustor, and the power turbine. In the compressor section, ambient air is drawn in and
compressed up to 30 times the ambient pressure and directed to the combustor section where fuel
is injected, ignited, and burned. The resultant gases are diluted with additional air from the
compressor section and are expanded through the power turbine section, which consists of a
series of rotors and stators to extract mechanical work via a shaft. A portion of the generated
shaft power is used to drive the internal compressor; the rest is directed to external load.
At natural gas compressor stations, natural gas processing facilities, and natural gas transmission
stations turbines are used mainly as prime movers to drive centrifugal compressors or generators.
Emissions from Natural Gas-Fired Turbines
Natural gas-fired turbines produce many of the same pollutants as SI-ICE which are emitted
from the exhaust, depending on the composition of the fuel used. In addition, PM emissions are
an issue for turbines due to the high exhaust flows. Since formaldehyde emissions from natural
gas-fired turbines are on the order of 7.10×10
-4
lb/MMBtu uncontrolled as per EPA’s AP-42
Emissions Factors, a 30,000 hp simple cycle turbine may emit approximately 0.6 tpy. However,

02/04/2017
Page
45
of
73
this size turbine is required to install an oxidation catalyst, which would also reduce
formaldehyde by 85-90%. Therefore, the Department has not established a formaldehyde limit
for simple cycle turbines. Natural gas is the primary fuel used by the natural gas industry and is
the only fuel authorized by GP-5.
Emission Control Technology
Several technologies may be used to control emissions from turbines. They primarily fall into
two categories, combustion control and post-combustion control.
Combustion Control
Control of combustion temperature has been the principal focus of combustion process control in
turbines. Combustion control requires tradeoffs – higher temperatures favor complete
consumption of the fuel and lower residual hydrocarbons and CO, but result in NO
X
formation.
Lean combustion dilutes the fuel mixture and reduces combustion temperatures and NO
X
formation.
Because the NO
X
produced by combustion turbines is primarily thermal NO
X
, reducing the
combustion temperature will result in less NO
X
production. Thus, the main strategy for NO
X
control is to control the combustion temperature. This is often done by using wet methods, such
as steam or water injection, or dry methods, such as lean combustion or two-stage combustion.
Steam or Water Injection
Steam or water injection has been demonstrated to effectively suppress NO
X
emissions from
turbines. The effect of steam or water injection is to increase the thermal mass by dilution and
thereby reduce peak temperatures in the flame zone. Steam or water is typically injected at a
water-to-fuel weight ratio of less than one. Depending on the initial NO
X
levels, such rates of
injection may reduce NO
X
by 60% or more. Both CO and VOC emissions are increased by
steam or water injection, and the level of increase will depend on the water-to-fuel weight ratio.
Dry Controls
Since thermal NO
X
is a function of both temperature and time, the basis of dry controls is to
either lower the combustor temperature using lean mixtures of air and fuel, fuel staging, or
decreasing the residence time of the combustor. A combination of these methods may also be
used to reduce NO
X
emissions.
Lean combustion involves increasing the A/F ratio of the mixture so that the peak and average
temperatures within the combustor will be less than that of the stoichiometric mixture, thus
suppressing thermal NO
X
formation. Introducing excess air not only creates a leaner mixture,
but also reduces residence time at peak temperatures.
Two-stage combustion can be broken down into lean/lean and rich/lean staging which both serve
to reduce NO
X
. In lean/lean staging, the combustor is a fuel-staged premixed combustor that
operates at an extremely lean A/F ratio. A small stoichiometric pilot flame ignites the premixed
gas to provide flame stability. Because the NO
X
emissions from the high temperature pilot flame
are insignificant and the combustor is designed to operate at lower flame temperatures and to
avoid localized “hot spots,” low NO
X
emission levels are achieved. In rich/lean staging, the
combustor is an air-staged premixed combustor where the primary zone is operated fuel rich and

02/04/2017
Page
46
of
73
the secondary zone is fuel lean. The fuel rich zone operates at an A/F ratio less than one, which
produces higher concentrations of CO and decreases the amount of NO
X
due to a lack of
available oxygen and lower flame temperatures. The exhaust from the primary zone is then
quenched and mixed with large amounts of air, creating a lean mixture which is pre-ignited and
introduced to the secondary zone where combustion is completed. The lower temperature and
lean mixture results in low NO
X
emission levels. Staged combustion is identified through a
variety of names, including Dry-Low NO
X
(DLN), Dry-Low Emissions (DLE), or SoLoNO
X
.
Post-Combustion Emission Reduction Technology for Turbines
Oxidation Catalyst (for CO and NMNEHC reduction)
Oxidation catalysts using platinum and palladium are effective for lowering CO, NMNEHC, and
formaldehyde levels in exhaust emissions from turbines. For this analysis, the Department has
determined that an oxidation catalyst is economically feasible for turbines if the cost per ton of
CO and NMNEHC removal is approximately $5,000; see Appendix C.
Selective Catalytic Reduction (for NO
X
reduction)
SCR is technically feasible on turbine exhaust streams, and the systems operate much like they
do on engine exhaust streams. Urea or ammonia is typically used in SCR systems that control
turbine NO
X
emissions, which also results in ammonia emissions. The Department evaluated the
potential NO
X
reductions and system costs for SCR and found that it is not cost effective. See
Appendix B for more details. For this analysis, the Department has determined that SCR is
economically feasible for turbines if the cost per ton of NO
X
removal is approximately $10,000.
Turbine Size Grouping
The Department chose to alter the turbine size groups for natural gas compressor stations, natural
gas processing plants, and natural gas transmission stations slightly. In the existing GP-5, the
turbines were placed into the following categories:
?
Greater than or equal to 1,000 bhp but less than 5,000 bhp;
?
Greater than or equal to 5,000 bhp but less than 15,000 bhp; and
?
Greater than or equal to 15,000 bhp.
In the new GP-5, the turbines were placed into the following categories:
?
Greater than or equal to 1,000 bhp but less than 5,000 bhp;
?
Greater than or equal to 5,000 bhp but less than 15,900 bhp; and
?
Greater than or equal to 15,900 bhp.
Turbine Emission Limits
New sources are required to control the emission of air pollutants to the maximum extent,
consistent with BAT as determined by the Department. The Department evaluated uncontrolled
emissions, control efficiency of various controls and associated costs, and stack test results for
turbines to establish BAT. In the following sections, all references to the pollutant
concentrations are given as ppm @ 15% O
2
.

02/04/2017
Page
47
of
73
Turbines Rated Greater Than or Equal To 1,000 bhp but Less Than 5,000 bhp
Vendor data for turbines greater than or equal to 1,000 bhp but less than 5,000 bhp gave
uncontrolled emission rates of 25 ppm NO
X
, 25 ppm CO, and 25 ppm THC as methane. This
was the basis of the BAT from the previous GP-5, and the THC emission rate was converted to
NMNEHC as propane, establishing an emission rate of 9 ppm NMNEHC as propane.
SCR cost estimations were based on vendor quotes, cited as from Vendor A and Vendor B. It
was assumed that the control efficiency for SCR is 90% for NO
X
. All oxidation catalyst cost
estimations were based on vendor data, with the costs quoted in 2007 dollars. The costs in the
analysis were then multiplied by the CPI of 1.16 for inflating 2007 dollars to 2016 dollars. It
was assumed that the control efficiencies for oxidation catalysts are 93% for CO and 50% for
NMNEHC. See Appendices B and C for the analyses.
Using the uncontrolled and BAT emission rates, the assumed control efficiencies, and assuming
full-year operation, the control cost for an oxidation catalyst for turbines greater than or equal to
1,000 bhp but less than 5,000 bhp is estimated between $8,142 and $13,810 per ton of pollutants
reduced. The Department determines that an oxidation catalyst is not economically feasible.
Therefore, BAT for turbines greater than or equal to 1,000 bhp but less than 5,000 bhp are
emission limits of 25 ppm for CO and 9 ppm for NMNEHC as in the previously issued GP-5.
Using the BAT emission rates, the assumed control efficiencies, and assuming full-year
operation, the control cost for SCR for turbines greater than or equal to 1,000 bhp but less than
5,000 bhp is estimated between $11,622 and $18,853 per ton of NO
X
reduced. The Department
determines SCR is not BAT for turbines greater than or equal to 1,000 bhp but less than
5,000 bhp because it is not economically feasible and the emission limit remains 25 ppm for
NO
X
as in the previously issued GP-5.
Turbines Rated Greater Than or Equal To 5,000 bhp but Less Than 15,900 bhp
Vendor data for turbines greater than or equal to 5,000 bhp but less than 15,900 bhp gave
uncontrolled emission rates of 25 ppm NO
X
, 25 ppm CO, and 25 ppm THC as methane. This
was the basis of the BAT from the previous GP-5 for CO and NMNEHC, and the THC emission
rate was converted to NMNEHC as propane, establishing an emission rate of 9 ppm NMNEHC
as propane. BAT for NO
X
was established as 15 ppm based on stack test results.
Using the uncontrolled and BAT emission rates, the assumed control efficiencies, and assuming
full-year operation, the control cost for an oxidation catalyst for turbines greater than or equal to
5,000 bhp but less than 15,900 bhp is estimated between $4,836 and $6,612 per ton of pollutants
reduced. However, according to stack test data, emission rates of 10 ppm CO and 5 ppm
NMNEHC are achievable without control. Using the stack test emission rates, the assumed
control efficiencies, and assuming full-year operation, the control cost for an oxidation catalyst
for turbines greater than or equal to 5,000 bhp but less than 15,900 bhp is estimated between
$11,082 and $15,153 per ton of pollutants reduced. Therefore, it is the Department’s
determination that turbines greater than 5,000 bhp but less than 15,900 bhp have dual BAT
criteria with emission limits of 10.00 ppm for CO and 5.00 ppm for NMNEHC uncontrolled, and
1.75 ppm for CO and 4.50 ppm for NMNEHC with control.
Using the uncontrolled emission rates, the assumed control efficiencies, and assuming full-year
operation, the control cost for SCR for turbines greater than or equal to 5,000 bhp but less than

02/04/2017
Page
48
of
73
15,900 bhp is estimated between $7,714 and $9,810 per ton of NO
X
reduced. Using the BAT
emission rates, the assumed control efficiencies, and assuming full-year operation, the control
cost for SCR for turbines greater than or equal to 5,000 bhp but less than 15,900 bhp is estimated
between $12,858 and $16,351 per ton of NO
X
reduced. Therefore, it is the Department’s
determination that turbines greater than 5,000 bhp but less than 15,900 bhp have dual BAT
criteria with emission limits of 15.00 ppm for NO
X
uncontrolled and 2.50 ppm for NO
X
with
control.
Turbines Rated Greater Than or Equal To 15,900 bhp
Vendor data for turbines greater than or equal to 15,900 bhp gave uncontrolled emission rates of
25 ppm NO
X
, 25 ppm CO, and 25 ppm THC as methane. This was the basis of establishing
oxidation catalysts as BAT from the previous GP-5 for CO and NMNEHC, and the THC
emission rate was converted to NMNEHC as propane, establishing an emission rate of 9 ppm
NMNEHC as propane. An alternative BAT of 10 ppm CO and 5 ppm NMNEHC based on stack
test results was offered in the previous GP-5. BAT for NO
X
was established as 15 ppm also
based on stack test results.
Using the uncontrolled emission rates, the assumed control efficiencies, and assuming full-year
operation, the control cost for an oxidation catalyst for turbines greater than or equal to
15,900 bhp is estimated between $3,576 and $4,321 per ton of pollutants reduced. Using the
alternative BAT emission rates, the assumed control efficiencies, and assuming a full year of
operation, the control cost for an oxidation catalyst for turbines greater than or equal to 15,
900 bhp is estimated between $8,243 and $9,903 per ton of pollutants reduced. Therefore, it is
the Department’s determination that turbines greater than 15,900 bhp have dual BAT criteria
with emission limits of 10.00 ppm for CO and 5.00 ppm for NMNEHC uncontrolled and
1.75 ppm for CO and 4.50 ppm for NMNEHC with control.
Using the BAT emission rates, the assumed control efficiencies, and assuming full-year
operation, the control cost for SCR for turbines greater than or equal to 15,900 bhp is estimated
to be less than $11,466 per ton of NO
X
reduced. The Department determines SCR is BAT for
turbines greater than or equal to 15,900 bhp. However, recently issued permits and plan
approval applications show turbines greater than or equal to 15,900 bhp are capable of achieving
9 ppm for NO
X
uncontrolled. Using the 9 ppm emission rate, the assumed control efficiencies,
and assuming full-year operation, the control cost for SCR for turbines greater than or equal to
15,900 bhp is estimated to be between $16,946 and $19,106 per ton of NO
X
reduced. Therefore,
it is the Department’s determination that turbines greater than 15,900 bhp have dual BAT criteria
with emission limits of 9.00 ppm for NO
X
uncontrolled and 1.50 ppm for NO
X
with control.
Table 8 - BAT Emission Limits for Existing Turbines
Turbine Rating
(bhp)
NO
X
(ppmdv @
15% O
2
)
CO
(ppmdv @
15% O
2
)
NMNEHC
(as propane)
(ppmdv @
15% O
2
)
Total PM
(lbs/MMBtu)
1,000 ≤ TR < 5,000
25.00
25.00
9.00
0.03
5,000 ≤ TR < 15,000
15.00
25.00
9.00
0.03
≥ 15,000
15.00
10.00
or
93% reduction
5.00
or
50% reduction
0.03

02/04/2017
Page
49
of
73
Table 9 - Proposed BAT Emission Limits for New Turbines
Turbine Rating
(bhp)
NO
X
(ppmdv @
15% O
2
)
CO
(ppmdv @
15% O
2
)
NMNEHC
(as propane)
(ppmdv @
15% O
2
)
Total PM
(lbs/MMBtu)
1,000 ≤TR < 5,000
25.00
25.00
9.00
0.03
5,000 ≤ TR < 15,900
15.00
Uncontrolled
or 2.50 with
Control
10.00
Uncontrolled
or 1.75 with
Control
5.00
Uncontrolled
or 4.50 with
Control
0.03
≥ 15,900
9.00
Uncontrolled
or 1.50 with
Control
10.00
Uncontrolled
or 1.75 with
Control
5.00
Uncontrolled
or 4.50 with
Control
0.03
In addition, the turbines shall comply with all applicable requirements specified in 40 CFR
Part 60, Subpart KKKK.
However, for all previous versions of the GP-5, the Department’s BAT requirements are more
stringent than those required under Subpart KKKK. Therefore, by complying with the
Department’s BAT requirements, the owner or operator of a turbine will be guaranteed
compliance with the applicable requirements of 40 CFR Part 60 Subpart KKKK.
Visible emissions shall not meet or exceed 10% opacity for a period or periods aggregating more
than three minutes in any one hour nor meet or exceed 30% opacity at any time.
Centrifugal Natural Gas Compressors
Like reciprocating natural gas compressors, centrifugal natural gas compressors are used to
increase the pressure of natural gas in a pipeline in order to take advantage of the property of
fluids moving from high-pressure to low-pressure areas. In a centrifugal compressor, however,
rotary motion from the prime mover is used to drive an impeller that imparts energy into the gas
which serves to increase its pressure. There can be multiple stages of impellers that can generate
a large change in pressure.
Like the reciprocating natural gas compressor, the centrifugal compressor has a shaft that must
be sealed to reduce wear and maintain gas pressure. These seals can be either dry or wet; the wet
seal uses an oil film in its operation. However, the oil in a wet seal collects natural gas, which
must be removed in order to maintain the seal effectiveness.
It is not typical for centrifugal compressors to be installed at an unconventional natural gas well
site. According to 40 CFR Part 60, Subparts OOOO and OOOOa, centrifugal compressors
located at well sites are not affected facilities. The Department is therefore not proposing to
authorize centrifugal compressors at unconventional well sites or remote pigging stations
through this General Permit.

02/04/2017
Page
50
of
73
Existing Centrifugal Natural Gas Compressors
The owner or operator of an existing wet seal centrifugal natural gas compressor shall continue
to comply with the 95% control and other applicable requirements specified in 40 CFR Part 60,
Subpart OOOO, which are detailed in the General Permit.
New Centrifugal Natural Gas Compressors
40 CFR Part 60, Subpart OOOOa has identical requirements to 40 CFR Part 60, Subpart OOOO
for wet seal natural gas compressors. Since the 95% control requirement would reduce methane
as well as VOC, no additional requirements were introduced.
The Department maintains that the use of a dry seal system is BAT and requires no other
requirements or limitations. The Department determines the requirements of 40 CFR Part 60,
Subpart OOOOa to be BAT for wet seal systems, except that the control efficiency of 98% is to
be applied consistent with the control efficiencies discussed in the previous section on enclosed
flares and other control devices. These requirements are detailed in the General Permit.
Natural Gas Fractionation Process Units
Condensates, or NGLs, are an important product of natural gas production. In much of the
production segment, the condensates are separated from the natural gas stream and stored in
tanks before eventually being shipped to a processing plant via truck. However, condensates are
still part of the natural gas stream and may fall out during transport in a pipeline. The fluid
buildup can cause flow problems, which are typically cleaned through a pigging operation. In
this case, the NGLs can be sent to a tank called a slug catcher which may be located at a
compressor station or processing plant. The liquids at a compressor station are also commonly
transported via a truck.
Natural gas fractionation is the process of separating the various hydrocarbons in NGLs by
extracting them in sequence in heated columns. Any methane, which is the lightest of the
hydrocarbons, remaining in the condensate is separated first, and put into the pipeline for
transmission and storage. Then ethane, propane, and butane are removed in turn and sent to their
respective storage tanks; butane may be further divided into isobutane and n-butane. The heavier
NGLs are called natural gasoline, and will typically be sent to another plant for further
refinement.
Potential emissions from natural gas fractionation units are from the process heaters and the
fugitive emissions associated with piping, valves, flanges, pump, compressors, and pressure
relief devices. Process heaters for fractionation columns eligible for authorization under this
general permit range in size up to 50 MMBtu/h. Combustion units rated less than 2.5 MMBtu/h
are exempt from plan approval by 25 Pa. Code §127.14(a). Under category number 39 of the Air
Quality Permit Exemption document, combustion units rated less than 10 MMBtu/h firing
natural gas supplied by an independent producer shall be given the same consideration given to
similarly sized sources that fire natural gas provided by a public utility. Even though the process
heaters are exempt, these sources will be listed in the permit for reference purposes and will
follow the required notification, reporting, and recordkeeping requirements. Emissions from
these exempt units will also be included in the facility emissions totals for tracking compliance
with the General Requirements (i.e., the 12 month rolling sum must remain below the major
source emissions thresholds) and in the annual emissions inventory.

02/04/2017
Page
51
of
73
Natural gas processing, which includes fractionation, are subject to federal requirements under
40 CFR Part 60, Subpart KKK for units which were constructed, reconstructed, or modified after
January 20, 1984 and on or before August 23, 2011. After August 23, 2011 and on or before
September 18, 2015, fractionation process units are subject to the requirements of 40 CFR
Part 60, Subpart OOOO. After September 18, 2015, fractionation process units are subject to the
requirements of 40 CFR Part 60, Subpart OOOOa. In all cases, the primary standards are
monitoring for equipment leaks using 40 CFR Part 60, Appendix A-7, Method 21.
Because potential emission sources associated with natural gas fractionation are addressed
individually, i.e. in the sections related to combustion units and fugitive emissions components, a
separate section for natural gas fractionation units was not incorporated into the General Permit.
Sweetening Units
Natural gas from some wells contains sulfur and carbon dioxide, which must be removed to
protect personnel, the environment, and equipment. Sulfur typically exists in natural gas in the
form of H
2
S, and natural gas where the H
2
S content exceeds 4 ppm is referred to as sour gas.
The process for removing H
2
S from sour gas is called sweetening the gas.
The primary process for sweetening sour gas is similar to the process of glycol dehydration. An
amine solution is used to remove the H
2
S by passing the sour gas through a tower where it
contacts the solution and is absorbed. There are two primary amine solutions used,
monoethanolamine (MEA) and diethanolamine (DEA). Either of these compounds, in liquid
form, will absorb sulfur compounds and CO
2
from the sour gas leaving the effluent gas virtually
free of these contaminants. Both MEA and DEA can be regenerated and the resultant gases can
be used to feed a Claus process, which involves using thermal and catalytic reactions to extract
elemental sulfur from the hydrogen sulfide solution.
11
Potential emissions from sweetening units are from the process heaters; the fugitive emissions
associated piping, valves, flanges, pump, compressors, and pressure relief devices; and the tail
gas of the Claus process. Process heaters for sweetening units are eligible for authorization
under this general permit up to 50 MMBtu/h in size. Combustion units rated less than
2.5 MMBtu/h are exempt from plan approval by 25 Pa. Code §127.14(a). Under Category
Number 39 of the Air Quality Permit Exemption document, combustion units rated less than
10 MMBtu/h firing natural gas supplied by an independent producer shall be given the same
consideration given to similarly sized sources that fire natural gas provided by a public utility.
Even though the process heaters are exempt, these sources will be listed in the permit for
reference purposes and will follow the required notification, reporting, and recordkeeping
requirements. Emissions from these exempt units will also be included in the facility emissions
totals for tracking compliance with the General Requirements (i.e., the 12-month rolling sum
must remain below the major source emissions thresholds) and in the annual emissions
inventory.
Sweetening units are subject to federal requirements under 40 CFR Part 60, Subpart LLL for
units which were constructed, reconstructed, or modified after January 20, 1984, and on or
before August 23, 2011. After August 23, 2011, and on or before September 18, 2015,
sweetening units are subject to the requirements of 40 CFR Part 60, Subpart OOOO. After
11
Web Site: NaturalGas.org,
http://naturalgas.org/naturalgas/processing-ng/#sulphur
.

02/04/2017
Page
52
of
73
September 18, 2015, sweetening units are subject to the requirements of 40 CFR Part 60,
Subpart OOOOa. The standards are for a target control efficiency for SO
2
emissions based on
sulfur production.
Because the Authorization to Use the GP-5 and GP-5A cannot be granted to facilities that
produce or process sour gas and as shown in the basic calculations in the section on Oxides of
Sulfur in General Methodology of Determining Best Available Technology, the SO
2
emissions
limits from the federal regulations were not included in the General Permits. Should a
sweetening unit be needed to remove excess CO
2
, the potential emissions sources from the
sweetening unit are addressed individually, i.e., in the sections on combustion units and fugitive
emissions components, and therefore a separate section on sweetening units was not incorporated
into the General Permit.

02/04/2017
Page
53
of
73
Appendix A – Average Gas Composition Analysis
Table 10 - Methane de Minimus Calculations
The general methodology for determining a methane de minimus is to calculate the amount of methane in a natural gas release relative to the amount of
VOC that reaches the VOC de minimis level. Using twelve different gas samples, the methane de minimis ranged from a minimum of 21.2 tpy to a
maximum of 1,615.3 tpy. The average of the twelve calculated de minimis is 714.9 tpy, which is 17,872 tpy of CO
2e
. This value is nearly one quarter of
the former 75,000 tpy CO
2e
major source threshold for greenhouse gases, which seems too high to be a de minimus value.
Therefore, the Department calculated an average gas composition from the twelve samples and followed the same methodology with a result of 191.6 tpy.
The Department decided to use 200 tpy for the methane de minimis level to provide some leeway due to the limited number of gas samples used in the
calculation, which is equivalent to 5,000 tpy CO
2e
. This is approximately 7% of the former greenhouse gas threshold and seems to be reasonable for a de
minimus value.
Average Gas
Composition
Donald R.
Bowser
#1M-207
Roundwood
Wyo
Martin
Sanders
1M
Liberty
Kenneth L.
Crosby
#1M-69
Petraitis
Boyanowski
Wyo
Jack
Wyo
Lopatofsky
Wyo
Fanclaire
Wyo
Susan
Sus
Delhagen
Sus
Methane
88.78%
63.08%
93.85%
83.21%
93.51%
70.94%
91.37%
93.90%
94.52%
95.41%
94.32%
95.49%
95.72%
Ethane
5.93%
7.91%
4.94%
4.98%
4.48%
17.15%
6.69%
4.77%
4.44%
3.86%
4.58%
3.71%
3.60%
Propane
0.88%
1.53%
0.36%
0.55%
0.36%
5.64%
0.73%
0.34%
0.26%
0.18%
0.31%
0.17%
0.15%
Butane
(iso- and n-)
0.27%
0.44%
0.05%
0.03%
0.05%
2.30%
0.13%
0.07%
0.04%
0.01%
0.05%
0.01%
0.01%
Pentane
(iso- and n-)
0.05%
0.09%
0.00%
0.00%
0.00%
0.49%
0.03%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
Hexane and
above
0.05%
0.00%
0.00%
0.00%
0.00%
0.62%
0.01%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
Carbon Dioxide
0.21%
0.16%
0.05%
0.30%
0.75%
0.17%
0.66%
0.06%
0.05%
0.06%
0.05%
0.07%
0.10%
Molecular
Nitrogen
3.85%
26.78%
0.74%
10.93%
0.41%
3.32%
0.41%
0.83%
0.69%
0.48%
0.65%
0.54%
0.42%
Sum of Parts
100.01%
99.99%
99.99%
100.00%
99.56%
100.63%
100.03%
99.97%
100.00%
100.00%
99.96%
99.99%
100.00%
VOC
(sum of propane
and above)
1.25%
2.06%
0.41%
0.58%
0.41%
9.05%
0.90%
0.41%
0.30%
0.19%
0.36%
0.18%
0.16%
VOC de
minimis (TPY)
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
Methane de
minimis (TPY)
191.6
82.7
618.0
387.4
615.8
21.2
274.1
618.4
850.7
1,355.8
707.4
1,432.4
1,615.3
Average of Columns (C through N) Methane de minimis (TPY)
714.9

02/04/2017
Page
54
of
73
Appendix B – SCR Cost Analysis for Engines and Turbines
Table 11 - SCR Cost Analysis for 1,380 HP Engine
(All dollar values in 2016 dollars)
Vendor A
@ 0.50 g/bhp-h
(2016 Quote)
Vendor B
@ 0.50 g/bhp-h
(2016 Quote)
Vendor C
@ 0.50 g/bhp-h
(2016 Quote)
Vendor A
@ 0.35 g/bhp-h
(2016 Quote)
Vendor B
@ 0.35 g/bhp-h
(2016 Quote)
Vendor C
@ 0.35 g/bhp-h
(2016 Quote)
SCR Purchased Equipment Costs
$107,000
$150,000
$107,000
$150,000
Reductant Tank Purchased Equipment Costs
$11,000
$30,000
$11,000
$30,000
Total Purchased Equipment Costs
$118,000
$180,000
$146,000
$118,000
$180,000
$146,000
Freight
$5,900
$9,000
$7,300
$5,900
$9,000
$7,300
Commissioning Costs
$0
$0
$11,500
$0
$0
$11,500
Total Indirect Installation Costs
$23,600
$36,000
$29,200
$23,600
$36,000
$29,200
Project Contingency
$21,240
$32,400
$26,280
$21,240
$32,400
$26,280
Total Plant Cost
$168,740
$257,400
$220,280
$168,740
$257,400
$220,280
Preproduction Cost
$3,375
$5,148
$4,406
$3,375
$5,148
$4,406
Inventory Capital - Initial Fill of Reductant
$2,500
$2,500
$2,500
$2,500
$2,500
$2,500
Total Capital Investment
$174,615
$265,048
$227,186
$174,615
$265,048
$227,186
Operating and Supervisory Labor Costs
$886
$886
$6,716
$886
$886
$866
Maintenance Cost
$2,619
$3,347
$3,408
$2,619
$3,347
$3,408
Reductant Consumption Cost
$18,540
$18,540
$24,090
$18,540
$18,540
$24,090
Annual Electricity Cost
$5,000
$5,000
$5,000
$5,000
$5,000
$5,000
Annual Catalyst Replacement Cost
$11,525
$11,525
$22,500
$11,525
$11,525
$22,500
Direct Annual Costs
$38,570
$39,297
$61,714
$38,570
$39,297
$55,864
Indirect Annual Costs
$16,482
$25,019
$21,445
$16,482
$25,019
$21,445
Total Annual Costs
$55,052
$64,316
$83,158
$55,052
$64,316
$77,308
TPY of NO
X
Emissions Reduced
5.99
5.99
5.99
4.19
4.19
4.19
Cost Per Ton
$9,189
$10,735
$13,880
$13,127
$15,336
$18,434

02/04/2017
Page
55
of
73
Table 12 - SCR Cost Analysis for 4,735 hp Engine
(All dollar values in 2016 dollars)
Vendor A
@ 0.50 g/bhp-h
(2016 Quote)
Vendor B
@ 0.50 g/bhp-h
(2016 Quote)
Vendor A
@ 0.35 g/bhp-h
(2016 Quote)
Vendor B
@ 0.35 g/bhp-h
(2016 Quote)
SCR Purchased Equipment Costs
$105,000
$225,000
$105,000
$225,000
Reductant Tank Purchased Equipment Costs
$20,000
$50,000
$20,000
$50,000
Total Purchased Equipment Costs
$125,000
$275,000
$125,000
$275,000
Freight
$6,250
$13,750
$6,250
$13,750
Commissioning Costs
$0
$0
$0
$0
Total Indirect Installation Costs
$25,000
$55,000
$25,000
$55,000
Project Contingency
$22,500
$49,500
$22,500
$49,500
Total Plant Cost
$178,750
$393,250
$178,750
$393,250
Preproduction Cost
$3,575
$7,865
$3,575
$7,865
Inventory Capital - Initial Fill of Reductant
$2,172
$2,172
$2,172
$2,172
Total Capital Investment
$184,497
$403,287
$184,497
$403,287
Operating and Supervisory Labor Costs
$1,771
$1,771
$1,771
$1,771
Maintenance Cost
$2,767
$6,694
$2,767
$6,694
Reductant Consumption Cost
$37,080
$37,080
$37,080
$37,080
Annual Electricity Cost
$5,000
$5,000
$5,000
$5,000
Annual Catalyst Replacement Cost
$23,050
$23,050
$23,050
$23,050
Direct Annual Costs
$69,668
$73,595
$69,668
$73,595
Indirect Annual Costs
$17,415
$38,067
$17,415
$38,067
Total Annual Costs
$87,083
$111,662
$87,083
$111,662
TPY of NO
X
Emissions Reduced
20.56
20.56
14.39
14.39
Cost Per Ton
$4,236
$5,432
$6,052
$7,760

02/04/2017
Page
56
of
73
Table 13 - SCR Cost Analysis for Turbines
Because the quotes furnished to the Department by the vendors were determined based on the assumption of 8,760 hours of operation, the Department
proposes to determine cost effectiveness of the control of NO
X
with an SCR system without regard to variability of hours of operation.
The Department is proposing to use the average of the total annual costs for each engine size as determined in Tables 1 and 2 above as a point on a line
to determine the cost effectiveness of SCR for sizes for which a quote was not obtained. The reason for using the average total annual cost versus the
average cost in dollars per ton is that the emission limit for NO
X
emissions for engines rated equal to or less than 500 hp is 1.0 g/bhp-h while for engines
(All dollar values in 2016 dollars)
Vendor A
1,590 HP
25 ppm NO
X
(2016 Quote)
Vendor A
30,000 HP
25 ppm NO
X
(2016 Quote)
Vendor B
30,000 HP
25 ppm NO
X
(2016 Quote)
Vendor A
30,000 HP
15 ppm NO
X
(2016 Quote)
Vendor B
30,000 HP
15 ppm NO
X
(2016 Quote)
Vendor A
30,000 HP
9 ppm NO
X
(2016 Quote)
Vendor B
30,000 HP
9 ppm NO
X
(2016 Quote)
SCR Purchased Equipment Costs
$514,600
$932,800
$2,000,000
$932,800
$2,000,000
$932,800
$2,000,000
Reductant Tank Purchased Equipment Costs
$15,000
$60,000
$60,000
$60,000
$60,000
$60,000
$60,000
Total Purchased Equipment Costs
$529,600
$992,800
$2,060,000
$992,800
$2,060,000
$992,800
$2,060,000
Freight
$26,480
$49,640
$103,000
$49,640
$103,000
$49,640
$103,000
Commissioning Costs
$0
$0
$0
$0
$0
$0
$0
Total Indirect Installation Costs
$105,920
$198,560
$412,000
$198,560
$412,000
$198,560
$412,000
Project Contingency
$95,328
$178,704
$370,800
$178,704
$370,800
$178,704
$370,800
Total Plant Cost
$730,848
$1,419,704
$2,945,800
$1,419,704
$2,945,800
$1,419,704
$2,945,800
Preproduction Cost
$14,617
$28,394
$58,916
$28,394
$58,916
$28,394
$58,916
Inventory Capital - Initial Fill of Reductant
$203
$1,408
$1,408
$1,408
$1,408
$1,408
$1,408
Total Capital Investment
$745,668
$1,449,506
$3,006,124
$1,449,506
$3,006,124
$1,449,506
$3,006,124
Operating and Supervisory Labor Costs
$6,716
$6,716
$6,716
$6,716
$6,716
$6,716
$6,716
Maintenance Cost
$11,185
$21,743
$45,092
$21,743
$45,092
$21,743
$45,092
Reductant Consumption Cost
$2,467
$17,150
$17,150
$17,150
$17,150
$17,150
$17,150
Annual Electricity Cost
$1,545
$17,501
$17,501
$17,501
$17,501
$17,501
$17,501
Annual Catalyst Replacement Cost
$20,741
$138,216
$138,216
$138,216
$138,216
$138,216
$138,216
Direct Annual Costs
$42,654
$201,326
$224,675
$201,326
$224,675
$201,326
$224,675
Indirect Annual Costs
$70,386
$136,823
$283,757
$136,823
$283,757
$136,823
$283,757
Total Annual Costs
$113,040
$338,149
$508,432
$338,149
$508,432
$338,149
$508,432
TPY of NO
X
Emissions Reduced
6.00
69.38
69.38
41.62
41.62
24.98
24.98
Cost Per Ton
$18,853
$4,874
$7,329
$8,124
$12,215
$13,538
$20,355

02/04/2017
Page
57
of
73
rate greater than 500 hp the emission limit for NO
X
is 0.5 g/bhp-h. This difference in emission limits would not be reflected using the average cost in
dollars per ton, as can be seen in Table 4 below.
Table 14 - Calculated Cost per Ton of NO
X
Reduced vs Average Cost per Ton NO
X
Reduced for Engines
The Department proposes that installing an SCR system on lean-burn engines rated at or above 1,875 HP is BAT if the uncontrolled emission rate is
0.50 g/bhp-h, resulting in an emission limit of 0.05 g/bhp-h. However, engine stack test data shows that engines in this size range are capable of
achieving an uncontrolled emissions rate of 0.35 g/bhp-h. As can be seen in Table 5 above, SCR is not economically feasible for an engine with
uncontrolled emissions rate of 0.35 g/bhp-h until an engine is rated at or above 3,000 HP. Therefore, the Department proposes dual BAT criteria for
lean-burn engines rated at or above 1,875 HP and less than 3,000 HP of 0.35 g/bhp-h uncontrolled or 0.05 g/bhp-h with control. The Department
proposes a BAT criterion for lean-burn engines rated at or above 3,000 HP of 0.05 g/bhp-h.
Following the methodology described above for engines with the turbine data results in cost per ton and average cost per ton as summarized in Table 5
below. The BAT emissions limits of 25 ppm for NO
X
for turbines below 5,000 bhp, of 15 ppm for turbines 5,000 bhp and above where used in the
analysis.
Engine HP
(All dollar values in
2016 dollars)
25
50
100
250
500
1,000
1,380
1,500
1,875
2,500
3,000
4,735
5,500
Average Total
Annual Costs
$53,926
$54,167
$54,650
$56,097
$58,509
$63,334
$67,000
$68,158
$71,777
$77,807
$82,632
$99,373
$106,754
TPY of NO
X
Emissions Reduced
0.22
0.43
0.87
2.17
4.34
4.34
5.99
6.51
8.14
10.85
13.02
20.56
23.88
Cost Per Ton
$248,427
$124,769
$62,940
$25,843
$13,477
$14,588
$11,183
$10,466
$8,818
$7,169
$6,344
$4,834
$4,471
Average Cost Per
Ton
$13,745
$13,698
$13,603
$13,320
$12,847
$11,901
$11,182
$10,955
$10,245
$9,063
$8,117
$4,834
$3,387
TPY of NO
X
Emissions Reduced,
Alternative BAT
5.70
7.60
9.12
14.39
16.71
Alternative Cost Per
Ton
$12,597
$10,241
$9,064
$6,906
$6,387

02/04/2017
Page
58
of
73
Table 15 - Calculated Cost per Ton NO
X
Reduced vs Average Cost per Ton NO
X
Reduced for Turbines using BAT Emissions
For turbines rated at 5,000 hp and above with uncontrolled emissions of 25 ppm NO
X
, SCR is cost effective. However, for turbines rated below
15,900 hp operating at 15 ppm NO
X
uncontrolled, SCR is not cost effective. Therefore, for turbines rated at or above 5,000 hp and below 15,900 hp, the
Department proposes an emission limit of 15 ppm NO
X
uncontrolled or 2.5 ppm NO
X
with control. For turbines rated at or above 15,900 hp, SCR is
cost effective even when the uncontrolled emission rate is 15 ppm NO
X
. However, a recently issued permit in New York establishes a NO
X
emission
limit of 9 ppm for two Solar Mars Turbines which according to the Solar Turbines website are rated at 15,900 bhp. A recent plan approval application
to the Department also proposes 9 ppm NO
X
for combustion turbines. Based on the availability of these low emission models, the Department proposes
an emission limit of 9 ppm NO
X
uncontrolled or 1.5 ppm NO
X
with control for turbines greater than 15,900 hp.
Turbine HP
(All dollar values in
2016 dollars)
1,000
1,590
3,000
5,000
6,130
7,500
11,150
15,900
17,500
20,000
25,000
30,000
Average Total
Annual Costs
$107,277
$113,040
$130,449
$153,621
$166,713
$182,586
$224,875
$279,908
$298,446
$327,411
$385,341
$423,291
TPY of NO
X
Emissions Reduced,
Uncontrolled
6.79
6.00
11.22
15.66
18.95
21.21
29.15
42.75
43.39
48.93
60.02
69.38
Uncontrolled Cost
Per Ton
$15,804
$18,853
$11,622
$9,810
$8,799
$8,611
$7,714
$6,547
$6,879
$6,691
$6,420
$6,101
Average Cost Per
Ton
$15,912
$15,933
$15,987
$16,062
$16,104
$16,155
$16,292
$16,470
$16,530
$16,624
$16,812
$16,999
TPY of NO
X
Emissions Reduced,
BAT
6.79
6.00
11.22
9.40
11.37
12.72
17.49
25.65
26.03
29.36
36.01
41.62
BAT Cost Per Ton
$15,804
$18,853
$11,622
$16,351
$14,665
$14,352
$12,858
$10,913
$11,466
$11,153
$10,701
$10,170
TPY of NO
X
Emissions Reduced,
Alternative BAT
15.39
15.62
17.62
21.61
24.98
Alternative Cost Per
Ton
$18,184
$19,106
$18,585
$17,832
$16,946

02/04/2017
Page
59
of
73
Appendix C – Oxidation Catalyst and NSCR Cost Analysis for Engines and Turbines
Unlike the methodology used in Appendix B for SCR, the Department relied on a memorandum from E
C
/
R
Incorporated dated June 29, 2010 to Melanie
King at EPA OAQPS/SPPD/ESG for engine emission control devices. The memorandum determined a linear equation for total annual cost for
oxidation catalysts for lean-burn engines. The Department then calculated the cost per ton based on weighted average emission rates and on BAT
emission rates.
Table 16 - Oxidation Catalyst Cost Analysis for Lean-Burn Engines - Weighted Average Emissions
Table 17 - Oxidation Cost Analysis for Lean-Burn Engines - BAT Emissions
The same memorandum also determined a linear equation for the total annual cost for NSCR for rich-burn engines. As for lean-burn engines, the
Department calculated the cost per ton for NSCR on rich-burn engines using the cost per ton based on weighted average emission rates and on BAT
emission rates.
Engine HP
(All dollar values in
2016 dollars)
25
50
100
250
500
1,000
1,380
1,500
1,875
2,500
3,500
4,735
5,500
Total Annual Costs
$3,906
$3,956
$4,058
$4,362
$4,869
$5,882
$6,653
$6,896
$7,656
$8,923
$10,950
$13,454
$15,005
TPY of CO
Emissions Reduced
17.84
35.69
5.82
14.54
8.68
17.37
23.96
26.05
34.19
45.58
63.82
86.34
100.29
TPY of NMNEHC
Emissions Reduced
0.02
0.05
0.29
0.72
1.21
2.41
3.33
3.62
7.24
9.65
13.51
18.27
21.22
Cost Per Ton
$219
$111
$664
$286
$492
$297
$244
$232
$185
$162
$142
$129
$123
Engine HP
(All dollar values in
2016 dollars)
25
50
100
250
500
1,000
1,380
1,500
1,875
2,500
3,500
4,735
5,500
Total Annual Costs
$3,906
$3,956
$4,058
$4,362
$4,869
$5,882
$6,653
$6,896
$7,656
$8,923
$10,950
$13,454
$15,005
TPY of CO
Emissions Reduced
0.43
0.87
1.74
4.34
1.74
3.47
4.79
5.21
6.51
8.68
12.16
16.45
19.10
TPY of NMNEHC
Emissions Reduced
0.02
0.05
0.34
0.84
0.60
1.21
1.66
1.81
2.26
3.01
4.22
5.71
6.63
Cost Per Ton
$8,523
$4,317
$1,956
$841
$2,081
$1,257
$1,030
$983
$873
$763
$669
$607
$583

02/04/2017
Page
60
of
73
Table 18 - NSCR Cost Analysis for Rich-Burn Engines - Weighted Average Emissions
Table 19 - NSCR Cost Analysis for Rich-Burn Engines - BAT Emissions
For lean-burn engines, using both the weighted average emissions and the BAT emissions, the cost effectiveness of control was shown for engines rated
at 50 bhp and above. For engines rated at 25 bhp, it was shown that control is not cost effective for engines emitting BAT emissions uncontrolled.
However, it is likely that engines rated at 25 bhp are unable to meet the BAT emissions without control; therefore, the Department determines that
oxidation catalysts are required for all lean-burn engines, regardless of rating.
Engine HP
(All dollar values in
2016 dollars)
25
50
100
250
500
1,000
1,380
1,500
2,500
3,500
4,735
5,500
Total Annual Costs
$6,494
$6,628
$6,895
$7,696
$9,032
$11,703
$13,733
$14,374
$19,716
$25,059
$31,657
$35,744
TPY of NO
X
Emissions Reduced
2.96
5.91
14.57
36.43
70.57
141.14
194.78
211.72
352.86
494.00
668.32
776.29
TPY of CO
Emissions Reduced
2.84
5.68
7.61
19.02
37.58
75.15
103.71
112.73
187.89
263.04
355.86
413.35
TPY of NMNE HC
Emissions Reduced
0.02
0.05
0.14
0.36
0.72
1.45
2.00
2.17
3.62
5.06
6.85
7.96
Cost Per Ton
$1,116
$569
$309
$138
$83
$54
$46
$44
$36
$33
$31
$30
Engine HP
(All dollar values in
2016 dollars)
25
50
100
250
500
1,000
1,380
1,500
2,500
3,500
4,735
5,500
Total Annual Costs
$6,494
$6,628
$6,895
$7,696
$9,032
$11,703
$13,733
$14,374
$19,716
$25,059
$31,657
$35,744
TPY of NO
X
Emissions Reduced
0.23
0.46
0.23
0.57
0.92
1.83
2.53
2.75
4.58
6.42
8.68
10.08
TPY of CO
Emissions Reduced
0.46
0.92
0.27
0.69
1.37
2.75
3.79
4.12
6.87
9.62
13.02
15.12
TPY of NMNEHC
Emissions Reduced
0.08
0.17
0.10
0.24
0.48
0.96
1.33
1.45
2.41
3.38
4.57
5.31
Cost Per Ton
$8,414
$4,294
$11,480
$5,126
$3,256
$2,110
$1,794
$1,727
$1,422
$1,291
$1,205
$1,172

02/04/2017
Page
61
of
73
For rich-burn engines, using the
BAT emissions shows that NSCR is
not cost effective below 250 hp;
however, NSCR was established as
BAT for rich-burn engines rated
greater than or equal to 100 bhp. For
rich-burn engines less than 100 hp,
both the weighted average emissions
and BAT emissions show NSCR as
cost effective except for 25 bhp
engines emitting at the BAT
emission rate. As stated above, it is
unlikely that this emission rate is
possible without the use of control;
therefore, the Department determines
that NSCR is required for all rich
burn engines, regardless of rating.
Unfortunately, a convenient analysis
for oxidation catalysts for turbines
was not found. Instead, the
Department relied on vendor quotes,
and performed an analysis similar to
the one for SCR in Appendix B.
Three independent quotes were given
for different sized turbines and those
formed the basis for the
extrapolation of total annual costs for
turbines of other sizes.
Table 20 - Turbine Characteristics and Emissions Data
Turbine A
Turbine B
Turbine C
Turbine D
Tubine E
Power (bhp)
1,590
6,130
11,150
15,900
30,000
Heat Rate (Btu/bhp-h)
10,370
8,500
7,190
7,395
6,360
Exhaust Flow (lb/h)
51,615
149,380
215,990
337,850
541,590
Exhaust Temperature (°F)
970
960
935
905
865
Heat Input (MMBtu/h)
16.49
52.11
80.17
117.58
190.80
Exhaust Flow (scfh)
720,747
2,085,927
3,016,063
4,717,703
7,562,708
Exhaust Flow (acfm)
32,411
93,145
132,308
202,505
315,113
NO
X
Emission Rate (ppm)
2.50E-05
1.50E-05
1.50E-05
9.00E-06
9.00E-06
NO
X
Emission Rate (lb/MMBtu)
0.09224
0.05534
0.05534
0.03321
0.03321
NO
X
Emission Rate (tpy)
6.66
12.63
19.43
17.10
27.75
CO Emission Rate (ppm)
2.50E-05
2.50E-05
2.50E-05
1.00E-05
1.00E-05
CO Emission Rate (lb/MMBtu)
0.05607
0.05607
0.05607
0.02243
0.02243
CO Emission Rate (tpy)
4.05
12.80
19.69
11.55
18.74
NMNEHC Emission Rate (ppm)
9.00E-06
9.00E-06
9.00E-06
5.00E-06
5.00E-06
NMNEHC Emission Rate (lb/MMBtu)
0.03179
0.03179
0.03179
0.01766
0.01766
NMNEHC Emission Rate (tpy)
2.30
7.25
11.16
9.09
14.76
y = 0.0061x + 12.568
0.00
50.00
100.00
150.00
200.00
250.00
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
Heat Rate
Heat Rate
Linear (Heat Rate)

02/04/2017
Page
62
of
73
Table 21 - Oxidation Catalyst Cost Data for Turbines
The uncontrolled emissions rates,
assume 25 ppm for CO and total
hydrocarbons, which when adjusted to
NMNEHC as propane becomes 9 ppm.
The BAT emissions rates, in ppm, were
taken from the previous GP-5. The
values for CO and NMNEHC are used
in the table below for the turbine
oxidation catalyst cost analysis.
(All dollar values in 2016 dollars)
6,130
15,900
30,000
Oxidation Catalyst Purchased Equipment Costs
$96,785
$205,918
$215,090
Direct Installation Costs (0.30PEC)
$29,035
$61,775
$64,527
Total Indirect Installation Costs (0.27PEC)
$26,132
$55,598
$58,074
Project Contingency (0.15(DIC+IIC))
$8,275
$17,606
$18,390
Total Capital Investment
$160,227
$340,897
$356,082
Operating and Supervisory Labor Costs
$18,889
$18,889
$18,889
Maintenance Cost
$2,904
$6,178
$6,453
Natural Gas Penalty
$12,553
$28,325
$45,964
Catalyst Disposal
$130
$338
$637
Annual Catalyst Replacement Cost
$14,204
$36,841
$69,512
Direct Annual Costs
$48,679
$90,570
$141,454
Indirect Annual Costs
$34,609
$60,854
$63,060
Total Annual Costs
$83,288
$151,424
$204,514
y = 4.9858x + 59939
$0
$50,000
$100,000
$150,000
$200,000
$250,000
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
Oxidation Catalyst Annual Costs
Oxidation Catalyst Annual Costs
Linear (Oxidation Catalyst Annual Costs)

02/04/2017
Page
63
of
73
Table 22
Oxidation Catalyst Cost Analysis for Turbines - Uncontrolled Emissions
Table 23 - Oxidation Catalyst Cost Analysis for Turbines - BAT Emissions
As can be seen in the two tables above, oxidation catalysts for turbines greater than or equal to 1,000 hp but less than 5,000 hp are cost ineffective; for
turbines greater than or equal to 5,000 hp, oxidation catalysts are cost effective. It is important to point out that for turbines less than 15,000 hp, the
uncontrolled emissions and the BAT emissions are identical, with identical results.
Turbine HP
(All dollar values in
2016 dollars)
1,000
1,590
3,000
5,000
6,130
7,500
11,150
15,900
17,500
20,000
25,000
30,000
Total Annual Costs
$64,925
$67,866
$74,896
$84,868
$83,288
$97,333
$115,531
$151,424
$147,191
$159,655
$184,584
$204,514
TPY of CO
Emissions Reduced
4.26
3.77
7.05
9.84
11.90
13.32
18.31
26.85
27.25
30.73
37.70
43.58
TPY of NMNEHC
Emissions Reduced
1.30
1.15
2.15
3.00
3.63
4.06
5.58
8.19
8.31
9.37
11.49
13.62
Cost Per Ton
$11,670
$13,810
$8,142
$6,612
$5,363
$5,600
$4,836
$4,321
$4,139
$3,981
$3,752
$3,576
Turbine HP
(All dollar values in
2016 dollars)
1,000
1,590
3,000
5,000
6,130
7,500
11,150
15,900
17,500
20,000
25,000
30,000
Total Annual Costs
$64,925
$67,866
$74,896
$84,868
$83,288
$97,333
$115,531
$151,424
$147,191
$159,655
$184,584
$204,514
TPY of CO
Emissions Reduced
4.26
3.77
7.05
9.84
11.90
13.32
18.31
10.74
10.90
12.29
15.08
17.43
TPY of NMNEHC
Emissions Reduced
1.30
1.15
2.15
3.00
3.63
4.06
5.58
4.55
4.61
5.20
6.38
7.38
Cost Per Ton
$11,670
$13,810
$8,142
$6,612
$5,363
$5,600
$4,836
$9,903
$9,486
$9,123
$8,599
$8,243
TPY of CO
Emissions Reduced,
Alternative BAT
3.93
4.76
5.33
7.32
TPY of NMNEHC
Emissions Reduced,
Alternative BAT
1.67
2.02
2.26
3.10
Cost Per Ton
$15,153
$12,291
$12,834
$11,082

02/04/2017
Page
64
of
73
For turbines greater than or equal to 15,900 hp, oxidation catalysts are cost effective if one considers the uncontrolled emission rates of 25 ppm CO and
9 ppm NMNEHC as propane as a baseline. However, if one considers the alternative BAT emissions rate established in the previous GP of 10 ppm CO
and 5 ppm NMNEHC, oxidation catalysts are cost prohibitive. Oxidation catalysts are also cost prohibitive for turbines rated greater than or equal to
5,000 hp but less than 15,900 hp with uncontrolled emissions rates of 10 ppm CO and 5 ppm NMNEHC, which is achievable according to stack test
data. Therefore, it is the Department’s determination that turbines greater than or equal to 5,000 hp have dual BAT criteria with BAT of 10 ppm CO
and 5 ppm NMNEHC uncontrolled and an emission rate of 1.75 ppm CO and 4.50 ppm of NMNEHC with control.

02/04/2017
Page
65
of
73
Appendix D – Cost Analysis for Combustion Control Devices
Table 24 - Combustion Control Device Cost Analysis
As can be seen from the table above, it is cost effective for a combustion control device that controls methane, VOC, and HAP based on the vendor
quote and assuming an emission rate at the de minimis threshold. In the sensitivity analysis, it is assumed that the annual costs scale relative to the
increased total capital investment. If methane is considered, it is cost effective for a combustion control device with a total capital investment of
$100,000 (and arguably much more). If methane is excluded from the cost per ton calculation, combustion control devices are cost effective up to a
total capital investment of $75,000.
(All dollar values in 2016 dollars)
Vendor A
Combustor
$19,186
Auto Ignitor
$1,740
Surveilannce System
$4,176
Total Purchased Equipment Cost
$25,102
Freight and Design
$1,740
Installation
$7,371
Total Capital Investment
$34,213
$40,000
$50,000
$75,000
$100,000
Pilot Fuel
$2,201
$2,573
$3,216
$4,824
$6,432
Maintenance Cost
$2,320
$2,712
$3,391
$5,086
$6,781
Data Management
$1,160
$1,356
$1,695
$2,543
$3,391
Direct Annual Costs
$5,681
$6,641
$8,302
$12,453
$16,603
Indirect Annual Costs
$3,756
$4,392
$5,490
$8,235
$10,979
Total Annual Costs
(7% Intrest, 15 Year Life)
$9,437
$11,033
$13,791
$20,687
$27,583
TPY of Methane
200.00
200.00
200.00
200.00
200.00
TPY of VOC
2.70
2.70
2.70
2.70
2.70
TPY of HAP
1.00
1.00
1.00
1.00
1.00
Cost Per Ton (95% Reduction)
$49
$57
$71
$107
$143
Cost Per Ton
(95% Reduction, Excluding Methane)
$2,685
$3,139
$3,924
$5,885
$7,847
Cost Per Ton (98% Reduction)
$47
$55
$69
$104
$138
Cost Per Ton
(98% Reduction, Excluding Methane)
$2,603
$3,043
$3,803
$5,705
$7,607
Sensitivity Analysis

02/04/2017
Page
66
of
73
Appendix E – LDAR Cost Analysis
ICF has conducted two analyses on LDAR, one for the Environmental Defense Fund (EDF) in March of
2014
12
and one for ONE Future Inc. in May of 2016
13
. In both analyses, they broke down the costs
associated with an LDAR program into an hourly cost number. This was an adaptation of the analysis
used by Colorado in their rulemaking. In the March 2014 analysis, the determination was that the total
cost as an hourly rate was $101.64 and that inspections would be performed on a quarterly basis. This
resulted in an LDAR cost that ranged between $2.15 and $7.60 per Mcf of methane reduced, which
approximates to between $95 and $336 per ton of methane reduced, when not counting the recovered
gas value. In the May 2016 analysis, the determination was that the total cost as an hourly rate was
$142.06 and that inspections would be performed annually. This resulted in an LDAR cost that ranged
between $1.41 and $6.94 per Mcf of methane reduced, which approximates to between $62 and
$306 dollars per ton of methane, when not counting the recovered gas value.
The Department conducted two independent LDAR cost analyses, the first using the ICF analyses as a
basis and a second based on two vendors’ quotes. The Department’s assumptions in the first analysis
included a semi-annual and quarterly survey interval for unconventional natural gas well sites and a
quarterly survey interval for natural gas compressor stations, processing plants, and transmission
stations. The semi-annual frequency is assumed to result in a 50% emissions reduction and the quarterly
frequency is assumed to result in a 60% emissions reduction. The Department favored the ONE Future
Equipment Costs as the basis of the cost analysis as it includes a high-flow system for leak
quantification in the analysis.
The two tables below represent the Department’s assumptions on frequency with the respective
assumptions on emissions and hours for each survey from the two ICF analyses.
Table 25 - LDAR Costs with ONE Future Equipment Costs and EDF Assumptions
12
Economic Analysis of Methane Emission Reduction Opportunities in the U.S. Onshore Oil and Natural Gas Industries, ICF
International on behalf of the Environmental Defense Fund, March 2014.
13
Economic Analysis of Methane Emission Reduction Potential from Natural Gas Systems, ICF International on behalf of
ONE Future Inc., May 2016.
Wellpads
Wellpads
Gathering
Processing
Transmission
Methane (Mcf/yr)
440
440
1,676
2,448
4,671
% Reduction
50%
60%
60%
60%
60%
Reduction (Mcf)
220
264
1,006
1,469
2,803
Hours each Inspection
2.7
2.7
8.0
8.0
8.0
Frequency per Year
2
4
4
4
4
Annual Inspection Cost
$767
$1,534
$4,546
$4,546
$4,546
Initial Set-Up
$77
$153
$455
$455
$455
Repair Labor Cost
$575
$1,151
$3,409
$3,409
$3,409
Total Cost per Year
$1,419
$2,838
$8,410
$8,410
$8,410
Cost of Reduction ($/Mcf)
$6.45
$10.75
$8.36
$5.73
$3.00
Cost of Reduction ($/ton)
$309.39
$515.65
$401.11
$274.62
$143.92

02/04/2017
Page
67
of
73
Table 26 - LDAR Costs with ONE Future Equipment Costs and Assumptions
As can be seen in the tables, the cost of reduction in both cases is well within the cost effective range.
The Department’s Independent LDAR Cost Analysis
Since the number of leaking components, amount of leakage from individual components, duration, and
frequency is uncertain at an individual facility, it is impractical to perform a generic cost-analysis for
LDAR requirements for static and rotating components (flanges, valves, joints, packing of compressor
rods, etc.) at well sites, natural gas compression, processing or transmission facilities.
However, the Department contacted representatives of two companies that offer services for LDAR for
the sources at well sites, natural gas compression, processing or transmission facilities, to determine the
estimated costs for leak detection, quantification, and repair tasks. Neither company offered quotes for
repair costs since the costs for repairs vary significantly, depending on which components are leaking
and what type of maintenance is required to repair the leaks. Therefore, the analysis is based only on the
quoted costs for methane gas leak detections and quantifications.
According to the representative of the first company, a dry natural gas well pad contains approximately
1,000 components. Wet gas well, compression, processing, or transmission facilities contain
approximately 2,000 components. Typically, it takes one person and one day (10 hours) to complete the
leak detection task using an OGI camera for a dry gas well pad. It takes one person two days to
complete the leak detection using an OGI camera for a wet gas well pad, natural gas compression,
processing, or transmission station. The company charges $75 per hour for manpower, $400 per day for
travel costs, $150 per day for camera rental, and $300 per day for gas leak detection device rental.
The company charges the same amount for labor for leak quantification, and assuming all components
are leaking it takes two people three days to quantify leaks from all 1,000 components at a dry well pad,
and two people six days to quantify leak from all 2,000 components at wet gas well pad, natural gas
compression, processing, or transmission facility. This yields a labor rate of 0.06 man-hours per
component that is leaking and in the analysis the Department will round up to the nearest full day.
The Department also contacted representatives of the second company, to determine their costs for
LDAR services, and they echoed the cost that was suggested by the first company. According to the
company representatives, typically it costs $750 - $1,500 for any well pad or natural gas compression
station for leak detection and quantification.
Based on the above information, the Department performed cost-effectiveness analysis for leak
detection and quantification requirements stated for sources located at well pads, natural gas
compression, processing, or transmission facilities. The costs for leak detection were assumed to be at
Wellpads
Wellpads
Gathering
Processing
Transmission
Methane (Mcf/yr)
3,057
3,057
3,605
5,986
3,605
% Reduction
50%
60%
60%
60%
60%
Reduction (Mcf)
1,529
1,834
2,163
3,592
2,163
Hours each Inspection
5.5
5.5
32.0
40.0
32.0
Frequency per Year
2
4
4
4
4
Annual Inspection Cost
$1,563
$3,125
$18,184
$22,730
$18,184
Initial Set-Up
$156
$313
$1,818
$2,273
$1,818
Repair Labor Cost
$1,172
$2,344
$13,638
$17,047
$13,638
Total Cost per Year
$2,891
$5,782
$33,640
$42,050
$33,640
Cost of Reduction ($/Mcf)
$1.89
$3.15
$15.55
$11.71
$15.55
Cost of Reduction ($/ton)
$90.71
$151.19
$745.92
$561.53
$745.92

02/04/2017
Page
68
of
73
the full quoted cost and the costs for leak quantification were based on the amount of time to survey 1%,
2%, and 3% of the components rounded up to the nearest full day for sources at dry gas well pads, wet
gas well pads, natural gas compression, processing, and transmission facilities.
According to Method 21, the detector must have a sampling rate between 3.53x10
-3
and 0.11 cf/min. At
the maximum sampling rate, a 500 ppm indication on a Method 21 detector would be approximately
equivalent to a 0.08 cf/day emission rate for a component if it is assumed that 100% of the leak is
captured. For leaks above 500 ppm, it is less likely that 100% of the leak is captured, but for purposes
of this analysis, it is assumed the detector does capture all of the emissions up to the 100,000 ppm level
for a calculated leak rate of 15.84 cf/day. For any leak above 100,000 ppm on a Method 21 device, a
high-flow sampler must be used for quantification. A high flow device has a sampling rate of 10 cf/min,
and can detect leaks between 0.05 cf/min and 8.00 cf/min. This results in leak emission rates of
72.00 cf/day and 11,520 cf/day, respectively.
Based on the several leak studies, one vendor informed the Department that the majority of leak
emissions come from a small percentage of “super-emitters.” The Department assumes the “super-
emitters” emit at the 11,520.00 cf/day rate and the total daily emission rate is based on the equation:
✇䰍
㨒㈇
찋븋
㬒㨒┇
氍猃猃眃球爃̀
㼇䈇
䀇㴇唇瀍
㨒㈇
Where:
E = Total Daily Emission Rate
P
SE
= Percentage of “Super-Emitters”
C
l
= Number of Leaking Components
P
E
= Percentage of Total Emissions Attributed to “Super-Emitters”
The following table outlines the calculated total daily emission rate for 1% of the total components at a
dry gas well site found leaking.
Table 27 - Super-Emitter Study Data and LDAR Costs
Study Author
Percent
“Super-
Emitters”
(P
SE
)
Percent
of Total
Emissions
(P
E
)
Calculated
Total Daily
Emission Rate
(E)
(cf/day)
Annual
Emissions
(Mcf/y)
Annual
Avoided
Emissions
(TPY)
Cost per
Ton of
Methane
Reduced
($/ton)
Brandt, et. al.
5.0%
50%
11,520.0
4,204.8
52.6
$228.13
Rella, et. al.
6.6%
50%
15,206.4
5,550.3
69.4
$172.82
22%
80%
9,642.6
14
3,519.5
44.0
$272.54
Clearstone Engineering
0.06%
58%
119.2
43.5
0.5
$22,047.27
British Columbia Oil & Gas
6.0%
80%
8,640.0
3,153.6
39.5
$304.17
The ONE Future emissions estimate from well sites is 3,057 Mcf/y and the EDF emission estimate from
well sites is 440 Mcf/y. Assuming the Department’s costs and that the emissions estimates are from 1%
of the total components, this results in costs of $313.78 per ton of methane reduced and $2,180.07 per
ton of methane reduced, respectively. The British Columbia Oil & Gas and the Rella, et. al. top 22% of
leaks are responsible for 80% of total emissions estimates are in line with ONE Future’s emissions
estimates from well sites.
14
The second line under Rella, et. al. is calculated assuming that 6.6% emit at the 11,520 cf/day rate and the remaining
15.4% emit at the 72.0 cf/day rate.

02/04/2017
Page
69
of
73
Brandt, et. al. and the Rella, et. al. top 6.6% are responsible for 50% of total emissions estimates are
nearly 80% higher than ONE Future’s emissions estimates. This serves to drive down the cost per ton of
methane reduced. The Clearstone Engineering emissions estimate is approximately 1.5% of the ONE
Future emission estimate; which serves to significantly drive up the cost per ton of methane reduced.
The Department therefore focused on the British Columbia and Rella, et. al. 22% cases in the cost
analysis.
Based on our independent analysis, the Department determines that quarterly LDAR is technically and
economically feasible for the control of methane emissions if over 2% of components are found to be
leaking. In this situation, the department considers quarterly LDAR to be BAT. However, if less than
2% of the components are leaking then the higher cost per ton of methane emissions reduced justifies
allowing the permittee to revert to the EPA’s semi-annual LDAR requirement.

02/04/2017
Page
70
of
73
Table 28 - LDAR Costs from Vendor Quotes
Table 29 - LDAR Cost Analysis Based on Rella et. al. and British Columbia Emissions Assumptions
Facility
Leak Detection
Duration (Days)
Labor Cost
Per Man-Hour
Number of
Hours/Day
Manpower
Travel Cost
($400/Day-Man)
Equipment Rental
(OGI)
($150/Day-Man)
Contingency
($200/Day-Man)
Frequency
per Year
Annual Cost for
Leak Detection
Wellpad (Dry Gas)
1
$75
10
1
$400
$150
$200
4
$6,000
Wellpad (Wet Gas),
Compressor Station,
Processing Plant, or
Transmission Station
2
$75
10
1
$800
$300
$400
4
$12,000
Facility
Percent
Leaking
Components
~No. of
Components
Leaking
Leak
Quantification
Duration
(Days)
Labor Cost
Per Man-Hour
Number of
Hours/Day
Manpower
Travel Cost
($400/Day-Man)
Equipment Rental
(Method 21)
($150/Day-Man)
Contingency
($200/Day-Man)
Frequency
per Year
Annual Cost for
Leak
Quantification
1.0%
10
1
$75
10
1
$400
$150
$200
4
$6,000
2.0%
20
1
$75
10
1
$400
$150
$200
4
$6,000
3.0%
30
1
$75
10
1
$400
$150
$200
4
$6,000
1.0%
20
1
$75
10
1
$400
$150
$200
4
$6,000
2.0%
40
1
$75
10
1
$400
$150
$200
4
$6,000
3.0%
60
1
$75
10
1
$400
$150
$200
4
$6,000
Wellpad (Wet Gas),
Compressor Station,
Processing Plant, or
Transmission Station
TOTAL ANNUAL COST FOR LEAK DETECTION
~No. of Components
1,000
2,000
TOTAL ANNUAL COST FOR LEAK QUANTIFICATION
Wellpad (Dry Gas)
Total Flow
Rate (CF/Day)
Undetected
Annual PTE
(CFY)
Avoided
Emissions
(CFY)
Mass of
Avoided
Emissions
(TPY)
$/ton
Total Flow
Rate (CF/Day)
Undetected
Annual PTE
(CFY)
Avoided
Emissions
(CFY)
Mass of
Avoided
Emissions
(TPY)
$/ton
1.0%
$12,000
9,643
3,519,549
2,111,729
44.03
$272.54
8,640
3,153,600
1,892,160
39.45
$304.17
2.0%
$12,000
19,285
7,039,098
4,223,459
88.06
$136.27
17,280
6,307,200
3,784,320
78.90
$152.09
3.0%
$12,000
28,928
10,558,647
6,335,188
132.09
$90.85
25,920
9,460,800
5,676,480
118.35
$101.39
1.0%
$18,000
19,285
7,039,098
4,223,459
88.06
$204.41
17,280
6,307,200
3,784,320
78.90
$228.13
2.0%
$18,000
38,570
14,078,196
8,446,918
176.12
$102.20
34,560
12,614,400
7,568,640
157.81
$114.06
3.0%
$18,000
57,856
21,117,294
12,670,376
264.18
$68.14
51,840
18,921,600
11,352,960
236.71
$76.04
Wellpad (Dry Gas)
Wellpad (Wet Gas),
Compressor Station,
Processing Plant, or
Transmission Station
TOTAL ANNUAL COST FOR LEAK DETECTION AND LEAK QUANTIFICATION
Facility
Percent
Leaking
Components
Total Annual
Cost
Rella et. al.
British Columbia Oil and Gas

02/04/2017
Page
71
of
73
Appendix F – Urbanized Areas and Urban Clusters from the 2010 Census
Albion, PA Urban Cluster (2010)
Allentown, PA--NJ Urbanized Area (2010)
Altoona, PA Urbanized Area (2010)
Ashland, PA Urban Cluster (2010)
Bedford, PA Urban Cluster (2010)
Bellefonte, PA Urban Cluster (2010)
Belvidere, NJ--PA Urban Cluster (2010)
Binghamton, NY--PA Urbanized Area (2010)
Blairsville, PA Urban Cluster (2010)
Bloomsburg--Berwick, PA Urbanized Area (2010)
Bonneauville, PA Urban Cluster (2010)
Bradford, PA--NY Urban Cluster (2010)
Brockway, PA Urban Cluster (2010)
Brookville, PA Urban Cluster (2010)
Burgettstown, PA Urban Cluster (2010)
Butler, PA Urban Cluster (2010)
Cambridge Springs, PA Urban Cluster (2010)
Chambersburg, PA Urbanized Area (2010)
Clarion, PA Urban Cluster (2010)
Cresson, PA Urban Cluster (2010)
Cumberland, MD--WV--PA Urbanized Area (2010)
East Liverpool, OH--WV--PA Urban Cluster (2010)
East Prospect, PA Urban Cluster (2010)
East Stroudsburg, PA--NJ Urbanized Area (2010)
Edinboro, PA Urban Cluster (2010)
Ellwood City, PA Urban Cluster (2010)
Emmitsburg, MD--PA Urban Cluster (2010)
Emporium, PA Urban Cluster (2010)
Erie, PA Urbanized Area (2010)
Everett, PA Urban Cluster (2010)
Fairdale, PA Urban Cluster (2010)
Franklin (Venango County), PA Urban Cluster (2010)
Greenville, PA Urban Cluster (2010)
Grove City, PA Urban Cluster (2010)
Hagerstown, MD--WV--PA Urbanized Area (2010)
Hanover, PA Urbanized Area (2010)
Harrisburg, PA Urbanized Area (2010)
Hazleton, PA Urbanized Area (2010)
Honesdale, PA Urban Cluster (2010)
Houtzdale, PA Urban Cluster (2010)
Huntingdon, PA Urban Cluster (2010)
Jersey Shore, PA Urban Cluster (2010)
Jim Thorpe, PA Urban Cluster (2010)
Johnstown, PA Urbanized Area (2010)
Kutztown, PA Urban Cluster (2010)
Lake Meade, PA Urban Cluster (2010)
Lancaster, PA Urbanized Area (2010)
Lebanon, PA Urbanized Area (2010)
Lewistown, PA Urban Cluster (2010)
Ligonier, PA Urban Cluster (2010)
Littlestown, PA Urban Cluster (2010)
Lock Haven, PA Urban Cluster (2010)
Lykens, PA Urban Cluster (2010)
Mansfield, PA Urban Cluster (2010)
Martinsburg, PA Urban Cluster (2010)
Masontown, PA Urban Cluster (2010)
Meadville, PA Urban Cluster (2010)
Mercer, PA Urban Cluster (2010)
Meyersdale, PA Urban Cluster (2010)
Mifflinburg, PA Urban Cluster (2010)
Mifflintown, PA Urban Cluster (2010)
Milford, NJ--PA Urban Cluster (2010)
Millersburg, PA Urban Cluster (2010)
Millsboro, PA Urban Cluster (2010)

02/04/2017
Page
72
of
73
Milton--Lewisburg, PA Urban Cluster (2010)
Monessen--California, PA Urbanized Area (2010)
Montgomery, PA Urban Cluster (2010)
Moscow, PA Urban Cluster (2010)
Mount Holly Springs, PA Urban Cluster (2010)
Mount Union, PA Urban Cluster (2010)
Muncy, PA Urban Cluster (2010)
Nanty-Glo, PA Urban Cluster (2010)
New Castle, PA Urban Cluster (2010)
New Freedom--Shrewsbury, PA--MD Urban Cluster (2010)
New Wilmington, PA Urban Cluster (2010)
North East, PA Urban Cluster (2010)
Northern Cambria, PA Urban Cluster (2010)
Oil City, PA Urban Cluster (2010)
Orwigsburg, PA Urban Cluster (2010)
Philadelphia, PA--NJ--DE--MD Urbanized Area (2010)
Philipsburg, PA Urban Cluster (2010)
Pine Grove, PA Urban Cluster (2010)
Pittsburgh, PA Urbanized Area (2010)
Portage, PA Urban Cluster (2010)
Pottstown, PA Urbanized Area (2010)
Pottsville, PA Urban Cluster (2010)
Punxsutawney, PA Urban Cluster (2010)
Quarryville, PA Urban Cluster (2010)
Reading, PA Urbanized Area (2010)
Reynoldsville, PA Urban Cluster (2010)
Ridgway, PA Urban Cluster (2010)
Roaring Spring, PA Urban Cluster (2010)
Saw Creek, PA Urban Cluster (2010)
Sayre--Waverly, PA--NY Urban Cluster (2010)
Scranton, PA Urbanized Area (2010)
Shamokin--Mount Carmel, PA Urban Cluster (2010)
Shippensburg, PA Urban Cluster (2010)
Sierra View--Indian Mountain Lake, PA Urban Cluster (2010)
Slippery Rock, PA Urban Cluster (2010)
Somerset, PA Urban Cluster (2010)
State College, PA Urbanized Area (2010)
Stewartstown, PA Urban Cluster (2010)
Susquehanna Depot, PA Urban Cluster (2010)
Titusville, PA Urban Cluster (2010)
Towanda, PA Urban Cluster (2010)
Treasure Lake, PA Urban Cluster (2010)
Tunkhannock, PA Urban Cluster (2010)
Tyrone, PA Urban Cluster (2010)
Union City, PA Urban Cluster (2010)
Uniontown--Connellsville, PA Urbanized Area (2010)
Waynesboro, PA--MD Urban Cluster (2010)
Waynesburg, PA Urban Cluster (2010)
Weirton--Steubenville, WV--OH--PA Urbanized Area (2010)
Williamsport, PA Urbanized Area (2010)
Williamstown, PA Urban Cluster (2010)
Youngstown, OH--PA Urbanized Area (2010)
Youngsville, PA Urban Cluster (2010)

02/04/2017
Page
73
of
73
Appendix G – Non-road Engine Standards
The emission limits in the following tables are expressed in g/bhp-h.
Table 30 - Non-Road Compression Ignition Engine Emission Standards
Tier 1
Tier 2
Tier 3
Tier 4i
Tier 4f
Model
Year
NO
X
HC
CO
PM
Model
Year
NO
X
HC
CO
PM
Model
Year
NO
X
HC
CO
PM
Model
Year
NO
X
HC
CO
PM
Model
Year
NO
X
HC
CO
PM
HP < 11
2000
7.83
5.97
0.75
2005
5.59
5.97
0.60
2008
5.59
5.97
0.30
2014
5.59
5.97
0.30
11≤HP < 25
2000
7.08
4.92
0.75
2005
5.59
4.92
0.60
2008
5.59
4.92
0.30
2014
5.59
4.92
0.30
25≤HP < 50
1999
7.08
4.10
0.75
2004
5.59
4.10
0.45
2008
5.59
4.10
0.22
2014
3.50
4.10
0.02
2013
3.50
3.73
0.02
50 ≤ HP < 75
1998
6.86
2004
5.59
3.73
0.30
2008
3.50
3.73
0.30
2013
3.50
3.73
0.02
2014
3.50
3.73
0.01
75≤HP < 100
2014
0.30
0.14
3.73
0.01
2014
0.30
0.14
3.73
0.01
100≤HP< 175
1997
6.86
2003
4.92
3.73
0.22
2007
2.98
3.73
0.22
2014
0.30
0.14
3.73
0.01
175≤HP< 300
1996
6.86
0.97
8.50
0.40
2003
4.92
2.61
0.15
2006
2.98
2.61
0.15
300 ≤HP< 600
1996
6.86
0.97
8.50
0.40
2001
4.77
2.61
0.15
2006
2.98
2.61
0.15
2014
0.30
0.14
2.61
0.01
2014
0.30
0.14
2.61
0.01
600 ≤HP≤750
1996
6.86
0.97
8.50
0.40
2002
4.77
2.61
0.15
2006
2.98
2.61
0.15
750 < HP ≤ 1205
2000
6.86
0.97
8.50
0.54
2006
4.77
2.61
0.15
2014
2.61
0.30
2.61
0.07
2014
Gen Set
0.50
0.14
2.61
0.02
1205 < HP
2014
Gen Set
0.50
0.30
2.61
0.07
2014
Other
2.61
0.14
2.61
0.03
2014
Other
2.61
0.30
2.61
0.07
Table 31 - Non-Road Spark Ignition Engine Emisison Standards
General Emission Standards
Severe-Duty Engine Emission Standards
NO
X
+ HC
CO
NO
X
+ HC
CO
Certification and Production-Line Testing
Tier 1
Model Year 2004-2006
2.98
37.29
2.98
96.94
Field Testing
4.03
37.92
4.03
96.94
Steady-State Testing
Tier 2
Model Year 2007-
2.01
3.28
2.01