United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/3-91-022a
November 1991
Air
Technical Guidance -
Stage II Vapor Recovery
Systems for Control of
Vehicle Refueling
Emissions at Gasoline
Dispensing Facilities
Volume I: Chapters
-------
EPA-450/3-91-022a
Technical Guidance -
Stage II Vapor Recovery Systems
for Control of Vehicle Refueling
Emissions at Gasoline
Dispensing Facilities
Volume I: Chapters
Emission Standards Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
November 1991
U.S. Environment.?! Pret-xibn Agency
Region 5, Library (P! .-*'•".
77 \''".zl Jackson 8cv>.
Chicago, IL bC^w.
-------
This report has been reviewed by the Emission Standards
Division of the Office of Air Quality Planning and
Standards, EPA, and approved for publication. Mention
of trade names or commercial products is not intended
to constitute endorsement or recommendation for use.
Copies of this report are available, as supplies permit
through the Library Services Office (MD-35), U.S.
Environmental Protection Agency, Research Triangle Park
NC 27711, (919) 541-2777, or for a nominal fee, from
National Technical Information Services, 5285 Port
Royal Road, Springfield VA 22161, (703) 487-4650.
11
-------
TABLE OF CONTENTS
VOLUME I
CHAPTER 1.0
CHAPTER 2.0
CHAPTER 3.0
INTRODUCTION
1.1
1.2
1.3
1.4
Background
Clean Air Act Requirements
Organization of Report
References
INDUSTRY DESCRIPTION
2.1
2.2
2.3
2.4
2.5
Industry Description
Industry Population and Size
Distribution
Model Plants
Summary
References
SOURCES OF EMISSIONS
3.1
3.2
3.3
3.4
3.5
3.6
General
Emission Sources
Factors Influencing Emissions
Emission Factor Calculations
Model Plant Emission Estimates
References
Page
1-1
1-2
1-3
1-9
1-11
2-1
2-1
2-4
2-28
2-30
2-31
3-1
3-1
3-6
3-11
3-15
3-29
3-32
CHAPTER 4.0
CONTROL TECHNOLOGY 4-1
4.1 Types of Stage II Systems 4-2
4.2 System Components 4-11
4.3 California Certification Program 4-33
4.4 In-Use Effectiveness 4-46
4.5 References 4-56
iii
-------
TABLE OF CONTENTS (Concluded)
CHAPTER 5.0
CHAPTER 6.0
STAGE II COSTS 5-1
5.1 Equipment, Installation
and Annual Costs 5-3
5.2 Model Plant Costs 5-16
5.3 comparison of Recent Cost studies 5-17
5.4 Current Costs of Stage II Systems 5-22
5.5 References 5-34
PROGRAM IMPLEMENTATION 6-1
6.1 Planning 6-2
6.2 Regulations 6-7
6.3 Permitting 6-12
6.4 Inspections 6-22
6.5 Summary 6-32
6.6 References 6-35
VOLUME II
APPENDIX A
APPENDIX B
APPENDIX C
APPENDIX D
APPENDIX E
APPENDIX F
APPENDIX G
APPENDIX H
APPENDIX I
APPENDIX J
APPENDIX K
LUNDBERG SURVEY INCORPORATED INDIVIDUAL
COUNTY SIZE DISTRIBUTION A-l
STAGE II FACILITY COSTS B-l
CALIFORNIA AIR RESOURCES BOARD
STAGE II (PHASE II) CERTIFICATION
TEST PROCEDURES C-l
CALIFORNIA AIR RESOURCES BOARD
EXECUTIVE ORDERS D-l
ILLUSTRATIVE EXAMPLE OF IN-USE
EFFICIENCY CALCULATION PROCEDURES E-l
STAGE II PROGRAM SUMMARIES F-l
PUBLIC AWARENESS INFORMATION G-l
STAGE II REGULATIONS H-l
PERMITTING INFORMATION 1-1
STAGE II TEST METHODS J-l
INSPECTION INFORMATION K-l
iv
-------
LIST OF FIGURES
Figure page
2-1 Gasoline Marketing in the United States 2-2
2-2 Comparison of Los Angeles Average Service
Station Size to MPSI Data 2-22
2-3 Comparison of EPA Nationwide, Sierra Los
Angeles, and Lundberg Retail Service
Station Size Distributions 2-25
3-1 Uncontrolled Service Station Operations 3-7
3-2 Controlled Service Station Operations
(Stage I and Stage II) 3-8
3-3 Region Boundaries 3-17
4-1 Vapor Balance System 4-3
4-2 Hasstech Assist System 4-5
4-3 Hirt Assist System 4-7
4-4 Healy Assist System 4-8
4-5 Amoco Bellowless Nozzle System 4-10
4-6 Example Balance Nozzles 4-12
4-7 Example Assist Nozzle 4-13
4-8 Example Bellowless Nozzle 4-14
4-9 High Hang Hose Configurations 4-25
4-10 Example Liquid Removal Device 4-27
4-11 Example Emergency Breakaway 4-28
4-12 Individual Vapor Balance System
Underground Piping 4-29
4-13 Manifolded Balance System Underground
Piping 4-31
4-14 Relationship of Inspection Frequency to
Program In-Use Efficiency 4-51
4-15 Relationship of Inspection Frequency to
Program In-Use Efficiency with Exemptions 4-54
5-1 Comparison of Installed Capital Costs
Lines Based on Data Point Averages 5-20
5-2 Comparison of Installed Capital Costs
Lines Based on Linear Regression 5-21
5-3 Comparison of Annual Costs Lines Based on
Data Point Averages 5-24
5-4 Comparison of Normalized Annual Costs
Lines Based on Linear Regression 5-25
-------
LIST OF TABLES
Table Page
1-1 OZONE NONATTAINMENT AREAS CLASSIFIED
MODERATE OR ABOVE 1-5
2-1 MONTHLY STATE GASOLINE CONSUMPTION FOR
1990 2-6
2-2 GASOLINE THROUGHPUT PERCENTAGES OF
NATIONAL TOTAL FOR OZONE NONATTAINMENT
AREAS CLASSIFIED MODERATE OR ABOVE 2-8
2-3 ESTIMATED GASOLINE CONSUMPTION BY STATE
FOR MODERATE AND ABOVE OZONE
NONATTAINMENT AREAS 2-9
2-4 ESTIMATED 1990 RETAIL SERVICE STATION
POPULATION 2-13
2-5 ESTIMATED PRIVATE SERVICE STATION
POPULATION 2-16
2-6 NATIONWIDE RETAIL SERVICE STATION
DISTRIBUTION ESTIMATED BY EPA 2-18
2-7 1990 MPSI MARKET SHARE BREAKDOWN 2-20
2-8 LOS ANGELES RETAIL SERVICE STATION
DISTRIBUTION REPORTED BY SIERRA RESEARCH 2-21
2-9 RETAIL SERVICE STATION DISTRIBUTION BASED
ON LUNDBERG DATA FROM 16 METROPOLITAN
AREAS 2-24
2-10 CONSUMPTION DISTRIBUTION FOR NATIONWIDE
AND METROPOLITAN AREA SCENARIOS 2-26
2-11 ESTIMATED PERCENTAGE OF RETAIL STATIONS
THAT ARE INDEPENDENTS BY THROUGHPUT
CLASSIFICATION 2-26
2-12 SERVICE STATION MODEL PLANTS AND
NATIONWIDE POPULATIONS 2-29
3-1 EXAMPLE COMPOSITION OF GASOLINE VAPORS 3-3
3-2 GASOLINE HAZARDOUS AIR POLLUTANT
VAPOR PROFILE 3-5
3-3 1992 AND BEYOND RVP LIMITS BY MONTH AND
BY GEOGRAPHIC LOCATION 3-13
3-4 MONTHLY AVERAGE DISPENSED LIQUID
TEMPERATURE 3-16
3-5 SEASONAL VARIATION FOR TEMPERATURE
DIFFERENCE BETWEEN DISPENSED FUEL
AND VEHICLE FUEL TANK (AT), *F 3-18
vi
-------
LIST OF TABLES (Concluded)
3-6 MONTHLY AND GEOGRAPHIC VARIATIONS IN
REFUELING EMISSION FACTOR 3-20
3-7 SUMMARY OF STAGEII/CONVENTZONAL REFUELING
SPILLAGE DATA 3-24
3-8 VOC EMISSIONS FROM REFUELING OPERATIONS
FOR SERVICE STATION MODEL PLANTS 3-30
4-1 SUMMARY OF CARS EXECUTIVE ORDERS CERTIFYING
SYSTEMS TO BE AT LEAST
95 PERCENT EFFICIENT 4-44
4-2 EFFICIENCY DECREASES ASSOCIATED WITH STAGE
II BALANCE SYSTEM DEFECTS 4-49
4-3 PERCENT CONSUMPTION EXCLUDED WITH VARIOUS
STAGE II EXEMPTION SCENARIOS 4-55
5-1 PURCHASE COSTS FOR VAPOR RECOVERY NOZZLES
AND REPLACEMENT PARTS (May 1991 Dollars) 5-6
5-2 TYPICAL VAPOR RECOVERY HOSE COSTS (May
1991 Dollars) 5-9
5-3 TYPICAL COSTS OF OTHER VAPOR RECOVERY
COMPONENTS (May 1991 Dollars) 5-9
5-4 PIPING COMPONENT DIFFERENCES BETWEEN
INDIVIDUAL AND MANIFOLDED BALANCE
SYSTEM 5-12
5-5 TYPICAL VAPOR PIPING COSTS FOR 65,000
GALLON PER MONTH SERVICE STATION 5-13
5-6 ACTIONS TAKEN IN RESPONSE TO FINDING A
LEAK IN AN UNDERGROUND TANK SYSTEM 5-15
5-7 SUMMARY OF STAGE II SYSTEM CAPITAL COST
ESTIMATES FROM ALL SOURCES 5-19
5-8 SUMMARY OF NORMALIZED STAGE II SYSTEM
ANNUAL COST ESTIMATES FROM ALL SOURCES 5-23
5-9 SUMMARY OF COST ITEMS CHANGED IN APPENDIX
B COST MODEL TO OBTAIN 1991 COSTS 5-27
5-10 1991 STAGE II BALANCE SYSTEM CAPITAL COST 5-29
5-11 1991 STAGE II BALANCE SYSTEM ANNUAL COST 5-30
5-12 COST EFFECTIVENESS OF 1991 STAGE II BALANCE
SYSTEMS 5-31
5-13 PROGRAM COST EFFECTIVENESS COMPARED TO
EXEMPTION LEVEL 5-33
6-1 SUMMARY OF STAGE II PROGRAM EXEMPTION
LEVELS AND COMPLIANCE SCHEDULES 6-13
6-2 PHASE II INSPECTION PROCEDURES 6-23
6-3 MASSACHUSETTS STAGE II VIOLATIONS 6-33
vii
-------
1.0 INTRODUCTION
The Clean Air Act Amendments (CAAA) of 1990 require the
installation of Stage II vapor recovery systems in many
ozone nonattainment areas and direct EPA to issue guidance
as appropriate on the effectiveness of Stage II systems.
This document provides guidance on the effectiveness of
Stage II systems and other Stage II technical information.
Stage II vapor recovery on vehicle refueling is an
effective control technology to reduce gasoline vapor
emissions that contain volatile organic compounds (VOC) and
hazardous air pollutants. Vehicle refueling emissions
consist of the gasoline vapors displaced from the automobile
tank by dispensed liquid gasoline. The stage II system
collects these vapors at the vehicle fillpipe and returns
them to the underground storage tank. Without vapor
recovery, the dispensing of gasoline causes the introduction
of fresh air into the storage tank. Liquid gasoline then
evaporates until liquid/vapor equilibrium is attained.
Stage II systems return saturated vapors to the storage tank
thus preventing this evaporation and actually saving
gasoline.
The purpose of this document is to provide information
and guidance to State and local agencies related to the
planning, permitting, and implementation of Stage II vapor
recovery programs. While the subject of enforcement is
introduced in this document, more detailed information and
guidance for enforcement programs are provided under
separate cover in the EPA's "Enforcement Guidance for Stage
II Programs" to be issued concurrently with this document.
1-1
-------
The information and guidance provided in this technical
document is not intended to establish a binding norm or a
final determination of issues or policies. Decisions on
issues and policies will be made during the development,
submittal, and review process on each individual State
Implementation Plan.
1.1 BACKGROUND
Stage II vapor recovery has been a part of VOC emission
control in California for some time. Since the introduction
of Stage II in California in the early 1970s, this program
has become one of California's major VOC control strategies.
Seventeen districts in California contain areas which are
classified nonattainment for ozone and have Stage II
programs that have been in effect for over a decade. It is
estimatjed that in California, Stage II vapor recovery
systems reduce hydrocarbon emissions by 48,000-56,000 tons
annually, and save 15-18 million gallons of gasoline.1'2 The
remaining districts in California have also recently adopted
hazardous air pollutant regulations reguiring Stage II vapor
recovery for control of benzene emissions.
Other areas of the country have also established Stage
II vapor recovery programs. The District of Columbia
implemented a Stage II program in the early 1980s and
Missouri adopted vehicle refueling regulations in the St.
Louis area in the late 1980s. In the late 1980s and early
1990s, several other States and local agencies adopted Stage
II programs. These agencies currently include New Jersey,
New York (New York City metropolitan area), Massachusetts,
Philadelphia, Washington, Oregon, and Dade County, Florida.
These programs range from ones that are well into the
implementation and enforcement period to those in the
initial stages. A number of additional areas are also
considering Stage II regulations.
1-2
-------
1.2 CLEAN AIR ACT REQUIREMENTS
The requirements in the CAAA of 1990 regarding Stage II
vapor recovery are contained in Title I: Provisions for
Attainment and Maintenance of National Ambient Air Quality
Standards. A key element of this title is that it
"classifies" areas with similar pollution levels. The
purpose of this classification system is to match pollution
control requirements with the severity of an area's air
quality problem. For ozone, there are five classes:
marginal, moderate, serious, severe, and extreme. Marginal
areas are subject to the least stringent requirements and
each subsequent classification is subject to more stringent
requirements. Areas in the higher classes must meet
requirements of all the areas in lower classifications plus
the additional requirements of their class.
Subject to the provisions of Section 202, Stage II
vapor recovery is required for moderate areas, and thus is
required for all areas classified as serious, severe, or
extreme. Section 182(b) of the CAAA of 1990 contains
requirements for moderate areas and section 182(b)(3)
specifically addresses gasoline vapor recovery.
(3) GASOLINE VAPOR RECOVERY.
(A) GENERAL RULE.-Not later than 2 years
after the date of the enactment of the Clean
Air Act Amendments of 1990, the State shall
submit a revision to the applicable
implementation plan to require all owners or
operators of gasoline dispensing systems to
install and operate, by the date prescribed
under subparagraph (B), a system for gasoline
vapor recovery of emissions from the fueling
of motor vehicles. The Administrator shall
issue guidance as appropriate as to the
effectiveness of such system. This
subparagraph shall apply only to facilities
which sell more than 10,000 gallons of
gasoline per month (50,000 gallons per month
in the case of an independent small business
marketer of gasoline as defined in section
325) .
(B) EFFECTIVE DATE - The date required
under subparagraph (A) shall be-
1-3
-------
6 months after the adoption
date, in the case of gasoline dispensing
facilities for which construction
commenced after the date of the
enactment of the Clean Air Act
Amendments of 1990;
(ii) one year after the adoption
date, in the case of gasoline dispensing
facilities which dispense at least
100,000 gallons of gasoline per month,
based on average monthly sales for the
2-year period before the adoption date?
or
(iii) 2 years after the adoption
date, in the case of all other gasoline
dispensing facilities.
Any gasoline dispensing facility described
under both clause (i) and clause (ii) shall
meet the requirements of clause (i).
(C) REFERINCI TO TERMS - For purposes of
this paragraph, any reference to the term
'adoption date1 shall be considered a
reference to the date of adoption by the
State of requirements for the installation
and operation of a system for gasoline vapor
recovery of emissions from the fueling of
motor vehicles.
Using nonattainment designations based on 1987-1989 design
values or a few areas based on 1988-90 design values, these
requirements would affect 56 metropolitan areas in the
United States. A breakdown of these areas by classification
is 32 moderate, 14 serious, 9 severe, and 1 extreme. The
areas are shown in Table 1-1.
In addition, Title 1, section 184, Control of
Interstate Ozone Air Pollution, creates an ozone transport
region comprised of the States of Connecticut, Delaware,
Maine, Maryland, Massachusetts, New Hampshire, New Jersey,
New York, Pennsylvania, Rhode Island, and Vermont, and the
CMSA that includes the District of Columbia.
1-4
-------
TABLE 1-1. OZONE NONATTAINMENT AREAS
CLASSIFIED MODERATE OR ABOVE
Extreme
Los Angeles-South Coast Air Basin, CA
Severe
Baltimore, MD
Chicago-Gary-Lake County, IL-IN
Houston-Galveston-Brazoria, TX
Milwaukee-Racine, WI
New York-N New Jer-Long Is.,
NY-NJ-CT
Philadelphia-Wilm-Trent,
PA-NJ-DE-MD
San Diego, CA
Southeast Desert Modified
AQMA, CA
Ventura Co, CA
Serious
Atlanta, GA
Baton Rouge, LA
Beaumont-Port Arthur, TX
Boston-Lawrence-Worcester
(E.MA), MA-NH
El Paso, TX
Greater Connecticut
Muskegon, MI
NJ
Portsmouth-Dover-Rochester,
NH
Providence (All RI), RI
Sacramento Metro, CA
San Joaguin Valley, CA
Sheboygan, WI
Springfield (Western MA), MA
Washington, DC-MD-VA
Moderate
Atlantic City,
Charleston, WV
Charlotte-Gastonia, NC
Cincinnati-Hamilton, OH-KY
Cleveland-Akron-Lorain, OH
Dallas-Fort Worth, TX
Dayton-Springfield, OH
Detroit-Ann Arbor, MI
Grand Rapids, MI
Greensboro-Winston Salem-High
Point, NC
Huntington-Ashland, WV-KY
Kewaunee Co, WI
Knox & Lincoln Cos, ME
Lewiston-Auburn, ME
Louisville, KY-IN
Manitowoc Co, WI
Miami-Fort Lauderdale-W. Palm
Beach, FL
Monterey Bay, CA
Nashville, TN
Parkersburg, WV
Phoenix, AZ
Pittsburgh-Beaver Valley, PA
Portland, ME
Raleigh-Durham, NC
Reading, PA
Richmond-Petersburg, VA
Salt Lake City, UT
San Francisco-Bay Area, CA
Santa Barbara-Santa
Maria-Lompoc, CA
St Louis, MO-IL
Toledo, OH
Source: 56 Federal Register 56692, 40 CFR 81, Air Quality
Designations; Final Rule. November 6, 1991.
1-5
-------
The requirements for this region also include provisions
related to Stage II, in section 184(b)(2).
(2) Within 3 years after the date of the
enactment of the Clean Air Act amendments of
1990, the Administrator shall complete a
study identifying control measures capable of
achieving emission reductions comparable to
those achievable through vehicle refueling
controls contained in section 182(b)(3), and
such measures or such vehicle refueling
controls shall be implemented in accordance
with the provisions of this section.
Notwithstanding other deadlines in this
section, the applicable implementation plan
shall be revised to reflect such measures
within 1 year of completion of the study.
In summary, all of the States in the transport region will
be required to implement Stage II controls or controls
determined by EPA to achieve comparable emission reductions,
Another portion of the Amendments with potential
impacts on the implementation of Stage II in moderate areas
is contained in Title 2: Provisions Relating to Mobile
Sources. Section 202, Control of Vehicle Refueling
Emissions, deals with the control of vehicle refueling
emissions using "onboard" systems. Onboard vapor control
systems consist of activated carbon canisters installed on
the vehicle to control refueling emissions. The carbon
canister system adsorbs the vapors that are displaced from
the vehicle fuel tank by the incoming liquid gasoline, and
subsequently purges these vapors from the carbon to the
engine when the engine is operating.
....The requirements of section 182(b)(3)
(relating to stage II gasoline vapor
recovery) for areas classified under section
181 as moderate for ozone shall not apply
after promulgation of such standards and the
Administrator may, by rule, revise or waive
the application of the requirements of such
section 182(b)(3) for areas classified under
section 181 as Serious, Severe, or Extreme
for ozone, as appropriate, after such time as
the Administrator determines that onboard
emissions control systems required under this
paragraph are in widespread use throughout
the motor vehicle fleet.
1-6
-------
This section has the effect of removing Stage II
requirements for moderate areas once onboard controls are
promulgated, and for the higher classified areas by EPA
rule, once onboard is in "widespread use".
The 1990 CAAA exempt, in section 182(b)(3), facilities
with gasoline throughputs of 10,000 gallons per month or
less and independent small business marketers (independents,
as defined in section 325 of the Clean Air Act as amended in
August 1977) with throughputs less than 50,000 gallons per
month. Section 325 has now been redesignated as section 326
by PL 98-213 and reads as follows:
Sec. 326. (a) The regulations under this
Act applicable to vapor recovery from fueling
of motor vehicles at retail outlets of
gasoline shall not apply to any outlet owned
by an independent small business marketer of
gasoline having monthly sales of less than
50,000 gallons. In the case of any outlet
owned by an independent small business
marketer, such regulations shall provide,
with respect to independent small business
marketers of gasoline, for a three-year
phase-in period for the installation of such
vapor recovery equipment at such outlets
under which such marketers shall have-
(1) 33 percent of such outlets in
compliance at the end of the first year
during which such regulations apply to such
marketers.
(2) 66 percent at the end of such second
year, and
(3) 100 percent at the end of the third
year.
(b) Nothing in subsection (a) shall be
construed to prohibit any State from adopting
or enforcing, with respect to independent
small business marketers of gasoline having
monthly sales of less than 50,000 gallons,
any vapor recovery requirements for mobile
source fuels at retail outlets. Any vapor
recovery requirement which is adopted by a
State and submitted to the Administrator as
part of its implementation plan may be
approved and enforced by the Administrator as
part of the applicable implementation plan
for that State.
(c) For purposes of this section, an
independent small business marketer of
1-7
-------
gasoline is a person engaged in "the
marketing of gasoline who would be required
to pay for procurement and installation of
vapor recovery equipment under section 324 of
this Act or under regulations of the
Administrator, unless such person-
(1)(A) is a refiner, or
(B) controls, is controlled by, or is
under common control with, a refiner,
(C) is otherwise directly or indirectly
affiliated (as determined under the
regulations of the Administrator) with a
refiner or with a person who controls, is
controlled by, or is under a common control
with a refiner (unless the sole affiliation
referred to herein is by means of a supply
contract or an agreement or contract to use
as a trademark, trade name, service mark, or
other identifying symbol or name owned by
such refiner or any such person), or
(2) receives less than 50 percent of his
annual income from refining or marketing of
gasoline.
For the purpose of this section, the term
"refiner" shall not include any refiner whose
total refinery capacity (including the
refinery capacity of any person who controls,
is controlled by, or is under common control
with, such refiner) does not exceed 65,000
barrels per day. For purposes of this
section, "control" of a corporation means
ownership of more than 50 percent of its
stock.
While this defines an independent marketer, it allows a
State or local agency to select an exemption level less than
50,000 gallons per month. A single exemption level approach
is currently taken by many regulatory agencies in their
Stage II programs.
There is another direct reference to Stage II vapor
recovery contained in the CAAA of 1977. This is section 324
regarding Cost of Emission Control for Vapor Recovery.
Sec. 324, (a) The regulations under this
Act applicable to vapor recovery with respect
to mobile source fuels at retail outlets of
such fuels shall provide that the cost of
procurement and installation of such vapor
recovery shall be borne by the owner of such
outlet (as determined under such
regulations). Except as provided in
1-8
-------
subsection (b), such regulations shall
provide that no lease of a retail outlet by
the owner thereof which is entered into or
renewed after the date of enactment of the
Clean Air Act Amendments of 1977 may provide
for a payment by the lessee of the cost of
procurement and installation of vapor
recovery equipment. Such regulations shall
also provide that the cost of procurement and
installation of vapor recovery equipment may
be recovered by the owner of such outlet by
means of price increases in the cost of any
product sold by such owner, notwithstanding
any provision of law.
(b) The regulations of the Administrator
referred to in subsection (a) shall permit a
lease of a retail outlet to provide for
payment by the lessee of the cost of
procurement and installation of vapor
recovery equipment over a reasonable period
(as determined in accordance with such
regulations), if the owner of such outlet
does not sell, trade in, or otherwise
dispense any product at wholesale or retail
at such outlet.
In summary, the clean Air Act and its 1990 Amendments
impose several direct requirements regarding Stage II vapor
recovery. The provisions in Title 1 will require that Stage
II controls be installed at all gasoline dispensing
facilities with throughputs above specified levels in
moderate, serious, severe, and extreme ozone nonattainment
areas, and Title II contains provisions which may relieve
the requirement for moderate and above areas if onboard
vehicle controls are promulgated. There are also direct
references that define independent marketers and describe
the party responsible for incurring the costs of vapor
recovery.
1.3 ORGANIZATION OP REPORT
The chief objective of this document is to provide
information pertaining to Stage II vapor recovery and
guidance to State and local agencies in the planning and
implementation of Stage II programs. Therefore, the report
1-9
-------
is organized in a manner that first provides an introduction
to Stage II vapor recovery and then emphasizes
implementation issues and potential problems.
Chapter 2 profiles the gasoline marketing industry,
with special consideration given to gasoline dispensing
facilities. Nationwide populations and size distributions
of these facilities are discussed as well as size
distributions representative of metropolitan areas. In
addition, model facilities are provided.
Chapter 3 discusses the sources of emissions at vehicle
refueling facilities, including the calculation of refueling
emission factors. This chapter also provides a discussion
of factors which influence refueling emissions. Emissions
are calculated for the model facilities described in
Chapter 2. Finally, emission factors are calculated on a
State basis taking into consideration RVP and temperature
differences across the nation.
Chapter 4 discusses vehicle refueling control
technology, both from a current and an historical basis. In
addition, a description of the California Air Resources
Board's (CARB) vapor recovery equipment certification
program is given which includes details of the certification
process and the certified equipment. Finally, the
effectiveness of the equipment is discussed, along with
program in-use efficiency.
Chapter 5 addresses the costs associated with Stage II
control. Equipment, installation, and maintenance costs are
discussed. Also, studies conducted in the St. Louis area
which include actual costs of stage II installations are
presented.
The final chapter is a guidance-oriented chapter which
uses the information presented in the earlier chapters. The
chapter discusses regulations and approaches to planning,
permitting, and enforcement, and is based on areas of the
country that have experience with Stage II vapor recovery
programs. It also addresses problems experienced by these
1-10
-------
agencies and suggested methods for others to use in avoiding
similar difficulties.
1.4 REFERENCES
1. McKinney, Laura. California Air Resources Board.
Gasoline Vapor Recovery Certification. (Presented
at the Air and Waste Management Association 83rd
Annual Meeting. Pittsburgh, PA. June 24-29,
1990).
2. Letter from Kunaniec, K., Bay Area Air Quality
Management District, to Shedd, S., U.S.
Environmental Protection Agency, Chemicals and
Petroleum Branch. July 31, 1991. Comments on
Preliminary technical guidance document.
1-11
-------
2.0 INDUSTRY DESCRIPTION
The purpose of this chapter is to define the industry
and facilities affected by a Stage II vapor recovery
program. The entire gasoline marketing industry is first
discussed, with special emphasis placed on the facilities
where gasoline is dispensed into vehicle fuel tanks (service
stations). Population and characteristics of the service
station industry are then addressed, including a discussion
of model dispensing facilities which may be used to
summarize the service station size distribution and
facilitate the estimation of environmental and economic
impacts.
2.1 INDUSTRY DESCRIPTION
The gasoline marketing industry includes many
components that move gasoline, from the refinery to the bulk
terminal and on to service stations. Gasoline produced by
refineries is distributed by a complex system comprised of
wholesale and retail outlets. Figure 2-1 depicts the main
elements in the marketing network. The flow of gasoline
through the marketing system is shown from the point of
production (the refinery), through bulk storage facilities
(bulk terminals), and finally to retail service stations or
private facilities where it is dispensed into vehicle fuel
tanks. Gasoline is often carried directly to the dispensing
facility from the bulk terminal; however, some gasoline
passes through intermediate storage and loading facilities
called bulk plants. The wholesale operations of storing and
transporting gasoline, including delivery to and storage in
2-1
-------
Imported or
Domestic Crude
Imported Gasoline
Barge/
Pipeline/
Tanker
Bulk Terminal
Tank
Truck
Bulk Plant
Service Station
Commercial, Rural
Accounts
Consumer
Figure 2-1. Gasoline Marketing In The United States
2-2
-------
a service station underground tank, are commonly called
Stage I operations. Vehicle refueling operations are
commonly termed Stage II.
Bulk gasoline terminals serve as the major distribution
point for the gasoline after it leaves the refinery.
Gasoline is most commonly delivered to terminals by
pipeline, but may also be transferred by ship or barge.
Gasoline is stored in large aboveground tanks and later
pumped through metered loading areas, called loading racks,
into delivery tank trucks. These tank trucks, in turn,
deliver product to various wholesale and retail accounts in
the marketing network.
Bulk gasoline plants are secondary distribution
facilities that typically receive gasoline from bulk
terminals transported by tank trucks, store it in
aboveground storage tanks, and subsequently dispense it into
smaller account trucks for delivery. Only a small portion
of the total gasoline is routed through bulk plants and much
of this eventually is delivered to private accounts and
small service stations.
Gasoline tank trucks are normally divided into
compartments with a hatchway at the top of each compartment.
Loading can be accomplished by top splash loading or
submerged fill through the hatch, or by bottom loading.
Either top or bottom loading can be adapted for vapor
collection. However, almost all gasoline is transferred
using bottom loading because of State vapor recovery
regulations and operating and safety advantages. The vapor
collection equipment on the truck is basically composed of
enclosed valves and piping that enable the vapors from the
compartment being filled to be transferred to the storage
tank being emptied (vapor balance) or to a vapor control
system.
Although the terms "service station", or "dispensing
facility", may be used to describe various types of
facilities, the term is used in this document to mean any
2-3
-------
site where gasoline is dispensed to motor vehicle fuel tanks
from stationary storage vessels. This includes both public
(retail) and private facilities. Miscellaneous retail
outlets that are considered service stations include
conventional service stations, convenience stores, and mass
merchandisers or "pumpers." Other facilities that may be
considered in this classification are marinas, parking
garages, and other similar facilities which sell gasoline to
the public.
Private facilities include those locations where
gasoline is dispensed into government agency (Federal,
military, State, and local) vehicles, fleet (auto rental,
utility companies, taxis, school buses, etc.) vehicles, and
trucking and local service vehicles. Other private
facilities include those that refuel farm equipment.
2.2 INDUSTRY POPULATION AND SIZE DISTRIBUTION
The volume of gasoline consumed and the number of
service stations in an area are important considerations in
assessing refueling emissions as well the potential emission
reductions, the economic impact, and even the overall
viability of a Stage II vapor recovery program. Also,
current and future trends are important in understanding the
industry and possible impacts. For example, the present
trend toward larger stations means that fewer stations and a
greater portion of the throughput would be subject to Stage
II controls. Also, the emergence of single nozzle multi-
product dispensers could greatly lessen the costs of Stage
II equipment and maintenance.
2.2.1 Gasoline Consumption
It is estimated by the Federal Highway Administration
that approximately 116 billion gallons of gasoline were
consumed in the United States in 1990.1 One can assume that
essentially this entire volume was eventually loaded into
vehicle fuel tanks, resulting in refueling VOC emissions.
Therefore, nationwide emissions from this source could have
2-4
-------
been almost 700,000 Mg of VOC/year, using a typical
uncontrolled refueling emission factor of 1,450 mg of
voe/liter of gasoline dispensed (discussed in Chapter 3).
As one would expect, gasoline consumption is directly
related to population. Therefore, States and areas with
high population density tend to show the highest gasoline
consumption figures. Monthly gasoline consumption by State
for 1990 is shown in Table 2-1.
It is estimated that over 40 percent of the gasoline in
the United States is consumed in ozone nonattainment areas
classified as moderate and above. This is due to the large
population density and vehicle traffic centered around the
metropolitan areas that traditionally have ozone attainment
problems. The percentage of the nationwide throughput for
each of the nonattainment areas shown in Table 1-1
represents is shown in Table 2-2. The estimated annual
gasoline consumption for ozone nonattainment areas by State
for 1990 is provided in Table 2-3. Ozone nonattainment area
consumption was estimated using county-to-State consumption
ratios calculated from EPA's 1985 NEDS gasoline consumption2
and the nonattainment counties are the final area
designations based on 1987-89 design values or 1988-90
design values for a few areas. These data show close to
half of the national throughput could be affected by Stage
II programs and that the impacts in serious, severe,
extreme, and possibly moderate areas could be considerable.
Since the recommended method used to calculate
refueling emissions is based on gasoline throughput,
accurate consumption estimates are critical. Gasoline
consumption data on a county basis are available from EPA's
National Air Data Branch. These data are calculated from
State gasoline consumption data provided by the Bureau of
the Census and apportioned to the county level using total
sales data. This approach has come under scrutiny, as the
relationship between gasoline consumption and total sales
2-5
-------
TABLE 2-1.
MONTHLY STATE GASOLINE CONSUMPTION FOR 1990
STATE
ALAIAMA
ALASKA
ARIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
DISTRICT OF COL.
FLORIDA
GEORGIA
W HAWAII
1 IDAHO
°* ILLINOIS
INDIANA
IOUA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA
HEM HAMPSHIRE
NiU JERSEY
NEW MEXICO
JAN
165,939
14,609
137,380
51,895
1,077.869
115,747
114.814
25,733
15,152
535,235
273.83*
31,198
38.27*
409,201
202,733
100,960
92.720
116,598
148,827
51,560
166,833
192,408
344,302
144,789
94,797
208,807
28,291
55,395
51,415
39,937
302,511
64,960
FEI
154,414
13,819
145,211
131.098
1,028,542
111,469
109,961
25,323
13,309
518,116
280,655
31.090
35,733
409,416
191,599
94,480
90,136
153,417
144,675
37,905
160,287
180,927
314,697
154,652
92,005
198,740
28,812
55,079
50,358
38,289
221,736
61,351
MAR
186,531
15.285
138,086
73,960
1,159,457
127,866
125,142
28,707
15,607
505,269
312,408
32,407
33,357
465.787
227,402
106,404
108,119
1(3,115
172,589
54,199
195,288
208,209
356,277
162,929
111,570
230,116
3*,*81
67,979
46,995
40.861
328,179
52,175
APR
in.esa
15,179
145,949
86,332
1,119,390
126,178
111,540
29,136
14,764
574,248
275,671
33,282
33,269
482,231
220,464
120,707
101,969
168,375
165,975
*2,*73
183,220
196,130
352,822
164,450
99,310
218,391
33,913
62.561
54,317
39,436
284,872
79,562
19-to DAtOliKt
MAY
189,481
51,944
112,623
162,742
1.138,520 1
141,839
126,939
50,027
14,604
525,085
330,619
33,420
41,609
411,797
233,439
123.052
112,759
170,5*1
179,173
52,990
183,323
212,614
390,339
188,586
118.895
252,839
38,926
70.617
52,645
42,612
239,093
73.298
COMMOT
JUN
183,308
26,920
148,067
111,956
,150,262
137,855
126,665
11,492
14,436
520,778
307,471
33,566
37,407
391,679
236,753
108.290
116,348
161,217
169,984
54,431
176,385
214,062
387,353
182,768
107,365
245,629
43.122
70,501
50,733
43,819
375,686
7*,773
m dooo
JUL
186,464
28,974
140,193
121,709
1.168,326
138,313
123,042
31,992
14,833
500,919
306,617
34.887
40,713
395,509
240,634
143,584
112,077
156,200
179,806
60,256
190,243
191,813
391,303
196,8*6
108,238
2*9,075
49,909
75,505
61,9*8
*7,177
309,278
73,021
CALi.au)
AUG
191,705
26,228
129,330
101,096
1,159,701
153,265
132,512
33,371
15,137
509,899
317,506
33.811
45,978
434.173
2*6,153
124,989
114,449
174,641
192,053
60,88*
189,391
2*1.377
412,546
177,129
115,830
250,767
36,779
74,732
58,064
50,800
247,482
73,841
SEP
161,861
23,926
129,330
115,78*
1.062,314
124,429
114,2(2
27,627
14,007
522.195
278,013
32,390
44,429
456,62*
215,356
95,928
96.113
146.076
162,263
50,859
171,300
193,879
337,977
194,750
94,174
220,082
39,779
63, *S?
62,802
41,555
291,073
66,990
OCI MOV DIC
177,862
20,266
152,291
100.401
1,105,746
133,247
120,320
28,956
14,650
465,047
299.760
32.637
56,150
478,223
235,317
142.902
104,344
157,958
169,473
52,286
183,326
204,467
372,412
180,346
108,075
235,178
36,779
67,221
53,205
43,637
348,928
66,520
173,545
18,780
144,595
48,154
1,068,403
116,404
120,031
27,572
14,650
117,679
294,924
31,5*6
42,606
414,694
221,785
95,5*8
100,321
1*8,210
161,399
42,695
17T.I76
199,116
363,925
165,939
106,249
228,086
36,779
64,291
53,384
40,969
295.584
66,798
172,296
18,203
135,21$
159,300
1,065,829
120.649
120,473
27,090
14,650
517.679
294,138
28,535
40,138
435,394
224,694
119,666
101,746
154,262
178,238
50,856
179,679
198,951
347,100
164,397
104,554
214,773
36,779
68,197
55,752
*1,045
295,584
58,140
TEAR
2,120,444
274,133
1,678,470
1,264,427
13,304,359
1,5*7,261
1,445,681
347,026
175,799
6,212,149
3,571,616
388,769
491,663
5,224.728
2,696,329
1,376,510
1,251,101
1,850,610
2,024,455
611,394
2, 15?, HI
2.433,953
4,371.053
2.077.581
1,265,062
2,752,483
444,349
795,515
651.818
510.137
3,547,006
811,429
-------
TABLE 2-1.
MONTHLY STATE GASOLINE
(CONTINUED)
CONSUMPTION FOR 1990
W
1
-J
STATE
NEU YORK
NOKTN CAROLINA
NORTH DAKOTA
OHIO
OKLAHOMA
OREGON
PEBNSYLVANIA
•WOE ISLAM
SOUTH CAROtiNA
SaiTH DAKOTA
TEHNESSEE
TEXAS
UTAH
VEiNONT
VIRGINIA
UASNIH6TON
WIT V1MIHIA
UttCMSIN
MTCNIHG
KAT10NUIDE
JAM
416.589
262,367
19,329
460,353
111.163
61, 6M
357,132
30,519
71,094
24.7*0
198,980
714,521
56,789
10,181
239,963
175.316
47,082
156,315
23,464
8,882,426
FEI
493,195
246,437
27,425
412,091
153,225
126,720
345,955
30,519
160,712
25,327
184,285
677.604
53,502
27.330
213,565
160,411
61,275
152,195
17,716
•,853,797
NAR
598,764
277,782
23,665
487,063
144,039
90,311
399,590
30,519
125,946
31.673
208,243
776,979
59,801
22.955
254,201
202,533
74.798
168.127
21,037
9,868.782
•««««*•«**«•
APR
445,437
288,658
32,824
492,083
139,945
136,149
384,101
32,143
178,544
29,080
241,257
741.879
58,438
20.836
270,852
185.078
73.300
166,178
19,320
9,7*9,256
1990 0* SOLI HE
HAT
532,657
292,9*1
29,421
516,493
1*8.103
119,973
*15,7«9
32,143
137,573
34,883
232,365
781.363
66,057
24.221
235.290
202,188
65.427
187,701
25,6*1
10,167,379 10
CONSUMPTION (1000
JUN JUL
S10.*63
290,722
33,532
517,808
159,462
103,397
41 1.489
32,143
135.151
37.487
211,570
789.124
65,571
24,976
302,746
200,590
76,811
187,206
21,597
,132,926
490,484
296,609
GALLONS)
AUG
591,929
307,581
36,277 38,267
508,673
1*6,128
129.325
409,257
32,602
134.837
42,168
243,6*9
769.82*
65,328
27.1*7
265,177
2U.681
80,280
200,411
24,123
10,186.384
539,737
159,863
1*6,157
429,245
32,602
134.837
42,585
217,877
781,771
71,697
28,852
273,380
220,004
69,914
206,212
21,843
10,439,972
SEP
509,934
260,703
29,783
468,174
134,289
125,192
381,578
32,602
134,837
32.606
224.550
694,567
60,361
23,325
227,481
193,794
77,823
171,249
21,8*3
9,458,255
OCT
509,934
279,940
29,500
499,189
133,199
115,784
404,551
31,755
134,837
32.098
224,501
720,121
61,132
25.770
264,404
195,956
63,746
172,991
21,843
9,869,181
NOV
509,934
268,306
28,294
383,635
143,742
88,293
400,717
31,755
134,837
31,258
202,317
704,669
55,636
22,994
257,535
185.935
55,234
177,632
21,843
9,347,103
DEC
509,934
265,453
23,883
480,482
139,334
123,641
394,488
31.755
134,837
29.991
224,043
707,070
60,032
22,054
223,248
174,645
69,608
176,929
21,843
9,557,272
TEAR
6,119,254
3,337,499
352,200
5.765,788
1,712,492
1.366,546
4,733,852
381,057
1,618,0*4
393,896
2,613,637
8,859.492
73*. 3*4
290,641
3,027,842
2,311,131
835,298
2,123,146
262,113
116,512,733
SOURCE: Federal Htghwty AdBlniitration, Monthly Gasoline Reports 1990, *s reported in 1991 NPN Fictbook
-------
TABLE 2-2. GASOLINE THROUGHPUT PERCENTAGES OF
NATIONAL TOTAL FOR OZONE NONATTAINMENT
AREAS CLASSIFIED MODERATE OR ABOVE
Percentage
of
Nonattainment Areas National
Throughput
Percentage
of
NonattainMnt Areas National
Throughput
Extreme
Los Angeles-South Coast Air Basin, CA 4.81
Severe
Baltimore, MD 0.99
Chicage-Gary-Lake County, IL-IH 2.52
Houston-Galveston-Brazoria, TX 1.64
Milwaukee-Racine, UI 0.52
New York-N New Jer-Long Is, NY-NJ-CT 4.97
Phtladelphia-Wilm-Trent, PA-NJ-DE-MD
Southeast Desert Modified AQMA, CA
San Diego, CA
Ventura Co, CA
Serious
Atlanta, GA 1.18
Baton Rouge, LA 0.2?
Beaumont-Port Arthur, TX 0.18
Boston-Laurence-Uorcester (E.HA), 2.40
MA-NH
El Paso, TX 0.17
Greater Connecticut 1.26
Muskegon, MI 0.05
Portsmouth-Dover-Rochester, NH
San Joaquin Valley, CA
Providence (All Rl), RI
Sacramento Metro, CA
Sheboygan, UI
Springfield (Western MA), MA
Washington, DC-MD-VA
Moderate
Atlantic City, NJ 0.12
Charleston, WV 0.12
Charlotte-Gastonis, NC 0.25
Cincinnati-Hani Iton, ON-KY 0.60
Cleveland-Akron-Lorain, OH 1.10
Dallas-Fort Worth, TX 1.63
Dayton-Springfield, OH 0.35
Detroit-Ann Arbor, MI 1.76
Brand Rapids, MI 0.25
Greensboro-Winston Salem-H Point, NC 0.30
Huntinston-Ashland, UV-KY 0.09
Keuaunee Co, UI 0.01
Knox & Lincoln Cos, ME 0.03
Lewiston-Auburn, ME 0.08
Louisville, ICY-IN 0.34
NanitOMoc Co, Wl 0.03
Miami-Fort Lauderdale-W. Palm Beach,
FL
Monterey Bay, CA
Nashville, TN
Parkersburg, WV
Phoenix, AZ
Pittsburgh-Beaver Valley, PA
Portland, ME
Raleigh-Durham, NC
Reading, PA
Richmond-Petersburg, VA
Salt lake City, UT
San Francisco-Bay Area, CA
Santa Barbara-Santa Maria-Lompoc, CA
St Louis, NO-II
Toledo, OH
1.91
a
0.86
0.23
13.64
0.13
0.98
0.35
0.73
0.00
0.31
1.12
9.13
1.52
0.23
0.37
0.07
0.84
0.86
0.17
0.26
0.13
0.07
0.30
2.16
0.13
1.06
0.20
15.50
Source: Honattairment designations from 56 ££ 56692 (See Table 1-1)
Gasoline consumption percentages estimated using 1985 NEDs fuel use report
* Gasoline consumption not reported because the consumption for this area and the LA South Coast Air Basin
consumption cited above overlap, and sufficient information is not in the database to allow proportion
this area's consumption from the LA consumption.
2-8
-------
TABLE 2-3. ESTIMATED GASOLINE CONSUMPTION BY STATE FOR
MODERATE AND ABOVE OZONE NONATTAINMENT AREAS
STATE
ALABAMA
ALASKA
ARIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
DISTRICT OF COL,
FLORIDA
GEORGIA
HAWAII
IDAHO
ILLINOIS
INDIANA
IOWA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA
NEW HAMPSHIRE
NEW JERSEY
NEW MEXICO
NEW YORK
NORTH CAROLINA
NORTH DAKOTA
OHIO
OKLAHOMA
OREGON
PENNSYLVANIA
RHODE ISLAND
SOUTH CAROLINA
SOUTH DAKOTA
TENNESSEE
TEXAS
UTAH
VERMONT
VIRGINIA
WASHINGTON
WEST VIRGINIA
WISCONSIN
WYOMING
TOTAL 1990
THROUGHPUT
(1000 gal) (1)
2,120,444
274,133
1,678,470
1,264,427
13,304,359
1,547,261
1,445,661
347,026
175,799
6,212,149
3,571,616
388,769
491,663
5,224,728
2,696,329
1,376,510
1,251,101
1,850,610
2,024,455
611,394
2,157,151
2,433,953
4,371,053
2,077,581
1,265,062
2,752,483
444,349
795,515
651,818
510,137
3,547,006
811,429
6,119,254
3,337,499
352,200
5,765,788
1,712,492
1,366,546
4,733,852
381,057
1,618,044
393,896
2,613,637
8,859,492
734,344
290,641
3,027,842
2,311,131
835,298
2,123,146
262,113
PERCENTAGE OF
THROUGHPUT IN
MODERATE AND ABOVE
OZONE NONATTAINMENT
AREAS (2)
OX
OX
57X
OX
94X
OX
100X
77X
100X
3 IX
40X
OX
OX
6 IX
12X
OX
OX
26X
14X
58X
86X
100X
55X
OX
ox
34X
OX
ox
ox
61X
98X
OX
49X
28X
OX
SOX
ox
ox
49X
100X
OX
OX
16X
45X
45X
OX
13X
OX
27X
3SX
OX
MODERATE AND
ABOVE OZONE
NONATTAINMENT
1990 THROUGHPUT
(1000 gal)
0
0
964,833
0
12,477,101
0
1,445,681
266,202
175,799
1,904,708
1,442,491
0
0
3,197,686
325,161
0
0
479,449
286,315
353,101
1,849,060
2,433,953
2,389,559
0
0
943,204
0
0
0
312,603
3,482,556
0
3,020,510
948,253
0
2,860,051
0
0
2,315,213
381,057
0
0
417,739
3,958,250
332,915
0
393,675
0
224,213
746,396
0
NATIONWIDE
SOURCES:
116,512,733 |
43X
50,327,735
(1) Federal Highway Administration, Monthly Gasoline Reports
As Reported in 1991 NPN Factbook
(2) Preliminary estimate based on 1987-89 design values
or 1988-90 design values for a few areas
2-9
-------
has not been well documented. EPA's Global Emissions and
Control Division, Air and Energy Engineering Research
Laboratory, in Research Triangle Park, NC, is studying this
issue in detail and plans to develop correlations with
other data such as population density, vehicle registration,
number of licensed drivers, highway usage, and many other
parameters, which will provide accurate estimates of
gasoline consumption on the county level.3
EPA's mobile source emission factor model, MOBILE4.1,
estimates refueling emission factors that are dependent on
either gasoline throughput or vehicle use, i.e. vehicle
miles travelled (VMT). As discussed in more detail in
Chapter 3, the emission factors are calculated in MOBILE4.1
using the same equation discussed in Section 3.4.1.
However, MOBILE4.1 also uses fuel economy information to
convert the emission factor from mass per gasoline
throughput to mass per VMT.
2.2.2 Service Station Population
While gasoline throughput, or consumption, is the
parameter used to calculate emissions, an estimate of the
number of facilities is necessary to help characterize the
affected community in more detail and to assess economic
impacts, both on industry and on regulatory agencies.
2.2.2.1 Retail Stations. A precise determination of
the number of retail service stations is very difficult.
The U.S. Census Bureau is the source usually relied upon for
information of this type. The Census Bureau provides
estimates of the number of retail service stations in the
Census of Retail Trade, but these data have limited
usefulness in defining the entire retail service station
industry. These reports are produced every five years and
have shown a steady and dramatic decrease in the number of
service stations. The reported service station population
has gone from 226,459 in 1972 to 114,748 in the most recent
1987 report.4 However, the definition of service station
used by the Census Bureau and the changing face of the
2-10
-------
industry make it difficult, if not impossible, to draw
conclusions from these estimates.
The Census Bureau defines as retail service stations
only those outlets that do 50 percent or more of their
dollar business in petroleum products. In 1972, this
provided a reasonably representative count of the retail
gasoline distribution facilities, as traditional service
stations accounted for the majority of retail outlets.
Today however, many facilities, such as convenience stores,
have large gasoline throughputs yet their sales from
gasoline may not total 50 percent of their sales due to the
wide variety of products offered.
An added problem with these census data is that they
consider only those stations that have payrolls. This
automatically excludes the privately owned and operated
family, or "Mom and Pop", facilities.
Another source of information traditionally used to
estimate retail service station population in the interim
period between Census Retail Trade Reports is "Franchising
in the Economy", a report formerly generated by the U.S.
Department of Commerce. This survey was discontinued in
January 1989, but was resumed by the International
Franchising Association, a private enterprise. These
reports also suffer from shortcomings as the definition of
service station is identical to that used by the Census
Bureau. The estimates by Franchising in the Economy place
the number of service stations in 1990 at 111,700.5
Franchising in the Economy does provide figures on
convenience store population. The 1988-1990 report accounts
for 17,000 stores. However, "National Petroleum News" (NPN)
refutes this number by estimating that there are as many as
80,000 convenience stores in business.6
After determining the need for a more accurate, current
estimate of retail gasoline dispensing facilities, NPN began
a vigorous nationwide survey. The results of this effort
were contained in the April 1991 issue of NPN.7 NPN
2-11
-------
embarked on this study by collecting the information on a
State basis, and allowing each State to be responsible for
its own statistics. Official figures for retail gasoline
station counts were not available for many States. The
study involved searching through motor vehicle department,
licensing department, and tax division records in more than
half of the States. NPN also contacted weights and measures
departments and key local trade associations. NPN estimated
that approximately 67 percent of the data obtained were
"hard" numbers; i.e., based on registration, licensing, and
tax division compilations. The remaining third were
obtained from unofficial estimates and, in a few cases, best
guess type estimates.
The results of this NPN study are provided in Table
2-4. As shown, the total retail service station population
in the nation is estimated to be 210,120. The NPN article
also discusses various methodologies which may be useful in
the determination of gasoline station population on a State,
regional, or local basis.
EPA has conducted several studies of the gasoline
marketing industry in connection with the development and
implementation of emission regulations. These studies
required estimates of the number of service stations. For
the most part, EPA has also relied on Census Bureau data as
the basis for its estimates. However, the Agency has long
recognized the shortcomings of these data and has attempted
to locate other sources of accurate information. EPA has
utilized service station retail population estimates of
approximately 211,000 in 1982,8 and 190,000 in 1984.9
In 1991, EPA is studying the hazardous air pollutant
(HAP) emissions from gasoline marketing sources in
accordance with Title III of the 1990 Clean Air Act
Amendments, including those from tank truck unloading at
service stations. During the search for information related
to nationwide service station population, EPA received
estimates of the current number of retail gasoline outlets
2-12
-------
TABLE 2-4. ESTIMATED 1990 RETAIL SERVICE STATION POPULATION
State
Number of
Stations
State
Number of
Stations
Alabama 6,500
Alaska 300
Arizona 4,010
Arkansas 3,764
California 13,800
Colorado 3,400
Connecticut 1,900
Delaware 450
Dist. of Columbia 134
Florida 10,152
Georgia 7,000
Hawaii 392
Idaho 1,123
Illinois 10,100
Indiana 4,500
Iowa 4,169
Kansas 3,062
Kentucky 2,446
Louisiana 6,600
Maine 700
Maryland 2,450
Massachusetts 2,500
Michigan 8,500
Minnesota 3,598
Mississippi 6,000
Missouri 7,200
Montana 1,400
Nebraska 3,000
Nevada 450
New Hampshire 1,050
New Jersey 3,860
New Mexico 2,066
New York 6,800
North Carolina 10,643
North Dakota 1,245
Ohio 6,205
Oklahoma 4,700
Oregon 2,165
Pennsylvania 6,000
Rhode Island 602
South Carolina 5,200
South Dakota 1,245
Tennessee 6,000
Texas 11,000
Utah 2,137
Vermont 856
Virginia 6,000
Washington 3,500
West Virginia 2,800
Wisconsin 5,074
Wyoming 1,372
NATIONWIDE TOTAL 210,120
Source: National Petroleum News, "Counting Procedure
How Retail Outlet Population is Greater Than
Expected," April 1991.
Shows
2-13
-------
from a number of sources. Independent estimates by both the
American Petroleum Institute (API)10 and Lundberg Survey,
Inc.11 placed the number of retail outlets at approximately
175,000.
The NPN estimates discussed earlier were considered.
However, EPA concluded that NPN article may slightly
overstate the retail population. Support for these
conclusions lies in the fact that Lundberg Survey recently
conducted a detailed survey of service stations in Arizona
that placed the population at 2,000, while the NPN article
estimated there are twice that number in the State.12 Also,
there are other questions raised by some of the NPN data,
one of which is seen when comparing State service station
population and gasoline throughput. For example, the NPN
numbers show that North Carolina has over two times as many
retail service stations as New York, while the gasoline
throughput is approximately 50 percent of New York's.
In lieu of any more precise or better supported number,
the 175,000 figure is being used for the 1990 nationwide
population of retail service stations in HAP analysis. This
is a significant increase in the total number from the
estimated 111,000 for 1989 in the Franchising in the Economy
data. This increase is primarily due to the inclusion of
"other" gasoline dispensing facilities not included in the
Census Bureau definition of service station.
While the nationwide estimate could be a point of
contention, there are essentially no affects of the
nationwide population for Stage II purposes. Since the
Stage II requirements contained in the 1990 CAAA are related
to ozone nonattainment areas only, the important service
station population figures are those for these nonattainment
areas. These nationwide estimates are included here to
provide States and local agencies with various information
related to retail service station population. These
agencies have the alternative to use any of this information
in estimating the population for their area.
2-14
-------
2.2.2.2 Private Stations. All of the estimates
discussed above are only for public, or retail facilities.
In addition to "public" outlets, there are a significant
number of "private" facilities. These outlets are
maintained by governmental, commercial, and industrial
consumers for their own fleet operations. Government
agencies with central garages are typically regional
locations for the postal service, Federal government
agencies, and State and county agencies. Other
miscellaneous facilities include utility companies, taxi
fleets, rental car fleets, school buses, and corporate
fleets. Estimated national population figures for private
facilities are shown in Table 2-513 The agricultural sector
of private outlets, including farms, nurseries, and
landscaping firms, are not included. In general,
agricultural outlets have throughputs less than the cutoff
levels. These private facilities are an important segment
of the industry and should be considered in population
estimates. The numbers shown in Table 2-5 were estimated in
1978. However, no more recent nationwide estimates have
been identified since this time.
2.2.2.3 Independents. One issue not addressed in any
of these estimates is the number of independent service
stations. As the Clean Air Act contains a different
exemption level for independents, it would be beneficial to
describe this segment of the industry. However, as
discussed in Chapter 1, the definition of "independent"
provided in the Clean Air Act is difficult to apply on a
quantitative basis. Also, the complex nature of service
station ownership and suppliers increases the difficulty of
a tally of independents. Estimates of relative percentages
of independent stations are discussed in the following
section.
2.2.3 Service Station Size Distribution
Not only is the number of facilities important to a
Stage II vapor recovery program, but estimates of the
2-15
-------
TABLE 2-5. ESTIMATED PRIVATE SERVICE STATION POPULATION8
"Private" Outlets
Government (Federal, military,
state, local) 85,450
Miscellaneous (auto rental,
utilities, others) 94,530
Trucking and Local Service 21,900
Taxis 5,380
School Buses 3,070
Total 210,330
8 Not including about 2.5 million agricultural outlets.
Source: "The Economic Impact of Vapor Recovery Regulations
on the Service Station Industry." EPA-450/3-78-029,
July 1978.
2-16
-------
relative sizes of facilities within the population are
needed for the cost analyses discussed later. The
parameters most useful to rank service stations are gasoline
throughput and the number of nozzles. This apportionment is
important for many reasons, but two principal ones are: (1)
to estimate the facilities which will be exempted, and (2)
to estimate the economic impacts of a regulation.
2.2.3.1 Retail Stations. The size distribution of
retail service stations according to gasoline throughput
used in the 1987 EPA Stage I study is given in Table 2-6.u
This size distribution, based on throughput, was used to
develop a national profile. The population is skewed toward
smaller stations, with over 75 percent having throughputs
less than 25,000 gallons per month.
Concerns have been raised regarding the applicability
of these estimates to larger metropolitan areas that are
typically nonattainment for ozone. In a 1988 report, "An
Analysis of Stage II and Onboard Refueling Emissions
Control" (Sierra Report),15 prepared by Sierra Research for
the Motor Vehicle Manufacturers Association, the
characteristics of the metropolitan service station
population are addressed. In this report, it is stated that
"EPA has ... failed to recognize that the average size of
gasoline stations in metropolitan nonattainment areas is
larger than the national average."
The Sierra Report contained a profile from Los Angeles
and compared it to the EPA estimates, to demonstrate the
difference in retail service station distribution for large
metropolitan areas. The use of Los Angeles data to
characterize all metropolitan areas in the United States is
questionable; however, Sierra did provide information
compiled by MPSI Americas, Inc. that suggests the Los
Angeles data are only slightly higher than other areas.
MPSI, Inc. of Tulsa, Oklahoma annually provides statistics
that are reported in the NPN Factbook. Among the statistics
are estimates of average facility gasoline consumption on a
2-17
-------
TABLE 2-6. NATIONWIDE RETAIL
SERVICE STATION DISTRIBUTION ESTIMATED BY EPA
Gasoline Throughput Range Percentage of Retail
(gallons/month) Service Stations
0 - 9,999 26
10,000 - 24,999 30
25,000 - 49,999 26.5
50,000 - 99,999 14
> 100,000 3.5
Source: "Draft RIA: Proposed Refueling Emission
Regulations for Baseline Motor Vehicles - Volume I
Analysis of Gasoline Marketing Regulatory
Strategies," EPA-450/3-87-001a.
2-18
-------
category basis. The categories are service stations,
pumpers, convenience food stores, and others. Overall
totals are also given. The MPSI summaries for 1990 as
contained in the 1991 NPN Factbook16 are shown in Table 2-7.
In order to validate the application of the Los Angeles data
to other areas of the country, Sierra used 1987 MPSI
information as reported in the 1988 NPN Factbook. Sierra
compared the average facility throughput for Los Angeles to
that reported by MPSI for 1987. The retail service station
size distribution from the Sierra Report for Los Angeles is
shown in Table 2-8, and the relationship of the Los Angeles
data to the 1987 MPSI data is illustrated in Figure 2-2.
The 1989 MPSI average service station size is also shown for
comparison in Figure 2-2.
EPA has obtained service station throughput data for
several metropolitan areas to verify the application of the
Los Angeles information presented in the Sierra Report to
metropolitan areas across the U.S. The data obtained were
compiled by the Lundberg Survey Incorporated17 and listed
gasoline stations and their associated gasoline monthly
volumes in gallons. There were approximately 11,000
individual service stations in the database which
represented 16 metropolitan statistical areas across the
United States. The areas included were:
Syracuse, NY Houston-Galveston-Brazoria, TX
Phoenix, AZ St. Louis, MI-IL
San Diego, CA Portland-Vancouver, OR-WA
Detroit, MI Milwaukee-Racine, WI
Lansing, MI New York-Newark-Long Island, NY-NJ-CT
Grand Rapids, MI Providence-Pawtucket-Fall River, MA-RI
£1 Paso, TX Madison, WI
Orlando, FL Santa Barbara-Santa Maria-Lompoc, CA
The service stations were placed into seven categories
according to monthly gasoline throughput. This was done for
2-19
-------
TABLE 2-7. 1990 MPSI MARKET SHARE BREAKDOWN
Northeastern Region
* X of Outlets
• % of Volume
* Avg. Monthly Volume
(Gallons)
Midwestern Region
» X of Outlets
• X of Volume
* Avg. Monthly Volume
(Gallons)
Sunbelt Region
• % of Outlets
• % of Volume
* Avg. Monthly Volume
(Gallons)
Western Region
* % of Outlets
* % of Volune
« Avg. Monthly Volume
(Gallons)
Total United States
« % of Outlets
* X of Volume
• Avg. Monthly Volume
(Gallons)
Service
Stations
60.6
54,7
62,611
35.9
28.2
59,220
22.0
23.4
55,613
45.6
42.6
70,428
38.4
36.4
62,479
Convenience
Pumpers Stores
22.3
39.2
121,861
43.7
63.0
108,706
34.5
57.8
101,853
34.2
50.0
127,931
33.2
52.5
112,230
6.3
3.6
39,847
9.2
6.0
42,642
33.2
15.7
28,735
12.4
5.4
38,252
18.3
8.5
32,220
Others
10.8
2.5
15,974
11.2
2.8
18,802
10.3
3.1
18,343
7.8
2.0
22,593
10.1
2.6
18,524
Total
100.0
100.0
69,360
100.0
100.0
74,782
100.0
100.0
58,798
100.0
100.0
82,356
100.0
100.0
69,036
Source: MPSI Inc., Tulsa, Oklahoma, reported in 1991 ttPN factbook.
2-20
-------
TABLE 2-8. LOS ANGELES RETAIL
SERVICE STATION DISTRIBUTION REPORTED BY SIERRA RESEARCH
Gasoline Throughput Range Percentage of Service
(gallons/month) Stations
0 - 9,999 12.9
10,000 - 24,999 8.0
25,000 - 49,999 21.8
50,000 - 99,999 35.2
> 100,000 22.0
Source: Sierra Research, "An Analysis of Stage II and
Onboard Refueling Emissions Control", November 30,
1988.
2-21
-------
80,000
60,000
40,000
20,000 H
0
Los Angeles
1987 MPSI
1990 MPSI
Figure 2-2. Comparison Of Los Angeles Average Service
Station Size To MPSI Data
2-22
-------
each county as well as an overall distribution for the
entire database. The overall distribution from these data
is shown in Table 2-9. As seen in the table, the
distribution is skewed toward the larger stations, just as
Sierra reported. More detailed breakdowns of the Lundberg
data are provided in Appendix A.
A side-by-side comparison of the EPA nationwide
distribution, the Sierra Los Angeles distribution, and the
Lundberg information distribution is provided in Figure 2-3.
These data indicate that the nationwide EPA distribution,
while accurate for nationwide analyses, may not be
appropriate for large metropolitan areas.
A comparison was also made between the consumption
distribution of the EPA nationwide facility distribution and
the metropolitan area distribution. Table 2-10 summarizes
this comparison. As would be expected from the facility
distribution, the throughput distribution in metropolitan
areas is skewed toward the larger throughput stations.
2.2.3.2 Private Stations. Based on information from
Arthur D. Little, Inc.18 and the U.S. Census Bureau,19 it was
previously estimated that approximately 90 percent of
private outlets have throughputs of less than 10,000 gallons
per month. In other analyses,20*21 1PA has used this figure
and distributed the remaining 10 percent in proportions
representative of the public service station distribution.
2.2.3.3 Independents, Previous 1PA analyses have
also estimated the relative percentages of retail facilities
that would be classified as "independent marketers'1 under
the Clean Air Act definition discussed in Chapter 1. Table
2-11 shows the relative percentages of retail stations that
are considered to be independents with the associated
throughput ranges.
These percentages were originally estimated during the
1984 Study based on information contained in EPA's report
"The Economic Impact of Vapor Recovery Regulations on the
Service Station Industry".22 This report categorized public
2-23
-------
TABLE 2-9. RETAIL SERVICE STATION DISTRIBUTION
BASED ON LUNDBERG DATA FROM 16 METROPOLITAN AREAS
Gasoline Throughput Range Percentage of Service
(gal Ions/month) stations
0 - 5,999 3.8
6,000 - 9,999 4.8
10,000 - 24,999 15.0
25,000 - 49,999 23.5
50,000 - 99,999 32.3
100,000 - 199,999 18.2
> 200,000 2.4
Source: Lundberg Survey, Incorporated.
2-24
-------
40
M
01
35
30
I
I
20
15
10
0
1
I
< 10,000
10,000-24,999
25,000-49,999
50,000-99,999
> 100,000
EPA Nationwide
Sierra Los Angeles
Lundberg Metropolitan
Figure 2-3. Comparison of EPA Nationwide, Sierra Los
Angeles, and Lundberg Retail Service Station Size
Distributions
-------
TABLE 2-10. CONSUMPTION DISTRIBUTION FOR NATIONWIDE
AND METROPOLITAN AREA SCENARIOS
Percent Consumption
Facility Throughput Range Nationwide Metropolitan
(gallons/month) Distribution Distribution
0 -
6,000 -
10,000 -
25,000 -
50,000 -
> 100
5,999
9,999
24,999
49,999
99,999
,000
4.7
4.1
17.8
27.5
27.2
18.8
2.4
0.4
5,0
12.4
29.1
50.6
TABLE 2-11. ESTIMATED PERCENTAGE OF RETAIL STATIONS THAT
ARE INDEPENDENTS BY THROUGHPUT CLASSIFICATION
Throughput Range Percentage of
(gall ons/month) independents
0 - 9,999 18%
10,000 - 24,999 31%
25,000 - 49,999 45%
50,000 - 99,999 39%
> 100,000 39%
2-26
-------
public service stations by company-controlled/company
operated, company- controlled/dealer operated, dealer
controlled/dealer operated, and convenience stores and
provided throughput distributions for each by direct
supplier and independent marketer/wholesaler. The
distributions in the Economic Impact Study were adjusted to
remove all convenience stores from independent marketers (it
is not expected that convenience stores obtain greater than
50 percent of sales from gasoline) and add all dealer-
controlled/dealer operated stations to independent
marketer/wholesaler. Based on the Census Bureau definition
of service station (greater than 50 percent of sales from
gasoline) and studies that estimate the total number of
public outlets that sell gasoline, an approximate ratio of
the Census population tot total population was estimated.
This ratio was approximately 2/3. The importance of this
ratio is that it indicates that approximately 1/3 of the
stations do not obtain over 50 percent of their sales form
gasoline. Therefore, the percentages obtained for
independent marketers were reduced by one-third.
2.2.4 Trends in the Service Station Industry
There are several trends in the service station
industry which could have an effect on a Stage II program.
Public acceptance of Stage II equipment is an important
aspect of any Stage II program. This is especially true in
light of the increase in the popularity of self-service type
stations. NPN reports substantial increases in the
percentage of self-service outlets across the country from
under 20 percent in 1975 to over 80 percent in 1989.23 A
similar trend is related to unattended gasoline stations.
This concept seems to be growing faster for commercial
fleets than for retail facilities. It is anticipated that
the number of convenience stores selling gasoline will
continue to increase, as well as the volume of gasoline sold
by these stores.
2-27
-------
As discussed in the previous section, the size of
service stations continues to rise. A steady increase in
the average facility gasoline throughput has been seen in
the last decade. The widespread popularity of dispensers
that allow the pumping of two or three gasoline products, or
"multiproduct dispensers" have allowed a station to have
more nozzles per station. However, the onset of dispensers
that have only one nozzle that can dispense multiple
gasoline products may cause a substantial decrease in the
number of nozzles per station.
Costs are discussed in Chapter 5, but one trend with
cost implications should be mentioned in this section. The
leaking underground storage tank (UST and LUST) programs,
depending upon the age and condition of the tank, require
replacement of tanks and/or piping. These programs could
affect Stage II programs in two different ways. First, if
the underground tanks and piping are replaced concurrently,
then the cost attributable to Stage II could be lessened.
Second, if these events do not occur simultaneously, then it
is possible that service station owners may be required to
initiate relatively major reconstruction more than once.
This issue is discussed in more detail in Chapter 5.
2.3 MODEL PLANTS
The development of typical, or model plants is a
technique often employed to assist in the determination of
impacts of a regulation during the planning stages. It is
preferable to develop several model plants to represent the
range of sizes of facilities present in the industry. The
distribution of facilities is applied to the model plants to
determine the relative percentage of facilities depicted by
each model plant.
In previous analyses,24'25 EPA has developed model plants
for the service station industry. The parameters selected
for the model plants are shown in Table 2-12.
2-28
-------
TABLE 2-12. SERVICE STATION MODEL PLANTS AND NATIONWIDE POPULATIONS
rsj
i
r\»
ID
Model Plant No. 1a 1b 2 3 4
Average Throughput 103 1 /mo 7.6 23.0 76.0 132.0 246.0
<103 gal/mo) (2) <6> (20) (35} (65)
Throughput Range 103 1 /mo 0-19 19-38 38-95 95-189 189-379
<103 gal/mo) (0-5) (5-10) (10-25) (25-50) (50-100)
Nuifcer of Nozzles
Single Dispensers 22 3 6 9
Nult {dispensers 66 6 12 18
Sources: 1987 Draft RIA.
5
700.0
(185J
>379
(>100>
15
30
-------
2.4 SUMMARY
It is important to develop an accurate characterization
of the industry that would be affected by a Stage II vapor
recovery regulation. This chapter has provided information
related to gasoline consumption, service station population,
size distribution, and model plants that may be useful to
agencies involved in these planning activities.
2-30
-------
2.5 REFERENCES
1. Federal Highway Administration, Monthly Gasoline
Reports 1990, as reported in 1991 National
Petroleum News (NPN) Factbook Annual Issue. June
1990.
2. 1985 NEDS Fuel Use Report. U.S. Environmental
Protection Agency National Air Data Branch.
Research Triangle Park, NC.
3. Memorandum from Norwood, P., Pacific Environmental
Services, Inc., to Shedd, S., U.S. Environmental
Protection Agency. April 11, 1991. Reporting on a
meeting with Larry Jones, EPA, AEERL, regarding
gasoline consumption project.
4. 1987 Census of Retail Trade. U.S. Department of
Commerce.
5. Franchising in the Economy 1988-1990. International
Franchising Association, Washington, D.C.
6. "Updated Survey Shows Fairly Static Growth In
Service Station Population", National Petroleum
News (NPN). April 1990.
7. "Counting Procedure Shows How Retail Outlet
Population Is Greater Than Expected", National
Petroleum News (NPN). April 1991.
8. Evaluation of Air Pollution Regulatory Strategies
for Gasoline Marketing Industry. U.S. Environ-
mental Protection Agency, Office of Air Quality
Planning and Standards and Office of Mobile
Sources. Publication No. EPA-450/4-84-012a. July
1984.
9. Draft Regulatory Impact Analysis: Proposed
Refueling Emission Regulations for Gasoline-Fueled
Motor Vehicles — Volume I - Analysis of Gasoline
Marketing Regulatory strategies. U.S. Environ-
mental Protection Agency. Office of Air Quality
Planning and Standards and Office of Mobile
Sources. Publication No. EPA-450/3-87-001a. July
1987.
10. Telecon. Thompson, S., Pacific Environmental
Services, Inc., with Peterson, B., American
Petroleum Institute. March 27, 1991. Number of
Service Stations.
2-31
-------
11. Telecon. Bollman, A., Research Triangle Institute,
with Keene, B., Lundberg Survey, Inc. April 19,
1991. Number of Public Service Stations.
12. Reference 11.
13. The Economic Impact of Vapor Recovery Regulations
on the Service Station Industry. U.S. Occupational
Safety and Health Administration, Washington, D.C.,
and U.S. EPA, Research Triangle Park, N.C.
Publication No. EPA-450/3-78-029. July 1978.
14. Reference 9.
15. Sierra Research, An Analysis of Stage II and
Onboard Refueling Emissions Control, prepared for
Motor Vehicle Manufacturers Association, Inc.
November 30, 1988.
16. "U.S. Regional and National Market Shares.", MPSI
Inc., Tulsa Oklahoma, as reported in 1990 National
Petroleum News (NPN) Factbook Annual Issue. June
1990.
17. Lundberg Survey, Inc. Census Data Gasoline
Throughput for U.S. EPA, June 28, 1989.
18. Reference 13.
19. 1977 Census of Retail Trade. U.S. Department of
Commerce.
20. Reference 8.
21. Reference 9.
22. Reference 13.
23. 1990 National Petroleum News (NPN) Factbook Annual
Issue. June 1990.
24. Reference 8.
25. Reference 9.
2-32
-------
3.0 SOURCES OF EMISSIONS
In this chapter, the emission sources at service
stations are described along with factors that affect the
rate at which emissions occur. In addition, emission
estimates or emission factors are presented that represent
emissions in different areas of the country. Emission rates
for different model facilities are presented to show how
total emissions vary by facility size and to characterize
rates for facilities throughout the country.
3.1 GENERAL
In virtually all cases in the gasoline marketing chain,
emissions of gasoline vapors are caused by the transfer of
liquid gasoline from one container (or tank) to another.
The liquid entering the fixed volume container displaces an
equal volume of gasoline vapor/air mixture to the
atmosphere. If the volume of vapor displaced from the
container equals the volume of liquid loaded into the
container, the ratio of vapor to liquid volume (V/L ratio)
is equal to 1.
However, the volume of vapors displaced often does not
equal the volume of liquid transferred. Temperature
variations between the liquid loaded and the vapors in the
tank can cause an expansion or contraction of the vapors
causing the V/L ratio to vary from 1. When warm liquid
enters a cool tank, the temperature in the tank increases
thereby increasing the volume of vapors in the tank and
increasing the volume of vapors displaced. This causes the
volume of displaced vapors to be greater than the volume of
3-1
-------
liquid loaded, resulting in a V/L ratio greater than 1.
This is called vapor growth.
The opposite occurs when the liquid entering the fixed
volume tank is cooler than the tank temperature. The cooler
temperature reduces the vapor volume displaced and the V/L
ratio is less than 1. This is called vapor shrinkage.
Vapor growth or vapor shrinkage can be a common occur-
rence when transferring liquids from service station
underground tanks containing liquid of relatively stable
temperature, insulated by the surrounding earth, into a
vehicle fuel tank at extreme temperatures caused by over-
road exposure to ambient conditions (fuel tanks very warm in
summer, very cold in winter), Because vapor growth and
vapor shrinkage occur so often, errors in emission estimates
can easily be encountered by simply assuming the volume of
vapors displaced equals the volume of liquid entering the
tank. Testing of these emission sources requires accurate
measurements of displaced volumes to calculate the mass of
emissions released.
Because the amount of emissions that occur is tied so
closely to the amount of liquid transferred into the tank or
container, emission factors are often expressed in terms of
mass emitted per volume of liquid loaded (i.e., pounds of
VOC per 1,000 gallons of liquid loaded or milligrams of VOC
per liter of liquid loaded).
Increased emphasis is being placed on the evaluation of
the emissions of hazardous air pollutants (HAPs). The CAAA
of 1990 specify 189 compounds that have been classified as
HAPs. Several of these HAPs are typically found in gasoline
vapors. Gasoline vapors are made up of a complex mixture of
compounds originating from the evaporation of liquid
gasoline.1 Table 3-1 shows an example mixture of compounds
found in displaced gasoline vapors. Several of these
compounds correspond with compounds found on the list of 189
3-2
-------
TABLE 3-1. EXAMPLE COMPOSITION OF GASOLINE VAPORS
Compound We ightPercent
N-Propane 4.6
Isobutane 19.0
N-Butane 21.4
Isopentane 28.3
N-Pentane 5.3
2-2-Dimethyl Butane 0.6
2-3-Dimethyl Butane 1.0
2-Methyl Pentane 4.0
3-Methyl Pentane 2.3
N-Hexane 1.l
3-3-Dimethyl Pentane 1.1
3-Methyl Hexane 0.7
Methyl Cyclopentane ' 1.2
Cis-2-Pentene 0.6
Benzene 0.7
Toluene 1.0
Other8 7.1
100
8 Other hydrocarbons with individual weight percent less
than 0.5.
Source: Furey, Robert and Nagel, Bernard. Composition of
Vapor Emitted From a Vehicle Gasoline Tank During
Refueling. SAE Technical Paper Series #860086,
February 1986.
3-3
-------
HAPs listed in Title III of the CAAA. Table 3-2 summarizes
the HAP compounds found in normal gasoline vapors and
indicates the percent of total emissions, on a weight basis,
that each HAP represents.2 These HAP emission rates were
calculated using liquid gasoline composition, Raoult's Law,
and gasoline vapor analyses. These values may not compare
exactly between Tables 3-1 and 3-2, since Table 3-1 is based
on one experimental sample group and the normal fuel profile
in Table 3-2 is based on a wide variety of samples.
The reformulated and oxygenated fuel requirements
contained in Title II of the CAAA will affect the HAP
content of gasoline. Also contained in Table 3.2 is an
estimate of a vapor profile for a reformulated gasoline.
Taken into account in this profile are the required
reductions in benzene and total aromatic content, the
addition of methyl tert butyl ether (MTBE) as an oxygenate,
and the reduction of all other components due to the
addition of a large volume of MTBE. HAP emissions from all
Stage I gasoline marketing sources (pipelines, terminals,
bulk plants, storage tanks, tank trucks, service station
underground tank loading) are being evaluated for regulation
under the National Emission Standards for Hazardous Air
Pollutant (NESHAP) program.
An interesting point is with regard to MTBE. MTBE is a
gasoline additive traditionally used in small amounts as an
octane booster. However, with oxygenated fuel requirements
contained in Title II of the 1990 Clean Air Amendments, the
addition of MTBE in gasoline will be widespread.
Approximately 15 weight percent MTBE in liquid gasoline is
needed to meet the 2.7 weight percent oxygen requirement for
carbon monoxide nonattainment areas, and 11 weight percent
to meet the 2.0 weight percent oxygen requirements for the
largest ozone nonattainment areas. This means that for
gasolines containing MTBE, 15 percent or more of gasoline
vapor could be made up of components listed by EPA as
hazardous pollutants.
3-4
-------
TABLE 3-2. GASOLINE HAZARDOUS AIR POLLUTANT VAPOR PROFILE
Hazardous Air Pollutant
HAP Content
HAP/VOC wt percentage ratio
Arithmetic Estimated
Average Normal Reformulated
Fuel Fuel
Hexane
Benzene
Toluene
2,2,4 Trimethylpentane
(iso-octane)
Xylenes
Ethylbenzene
Naphthalene
Cumene
MTBE
TOTAL HAPS*
1.6
0.9
1.3
0.8
0.5
0.1
0.5
0.1
4.8
1.4
0.4
1.1
0.7
0.4
0.1
0.0
0.0
8.7
13
8 Columns do not add to totals. Total HAPs as well as
individual HAPs were calculated for each data point in
the normal fuel analysis, and thus the totals are not
simply sums of the individual components. Adjustments
were made to this normal fuel based on the reformulated
gasoline requirements to predict a reformulated profile.
Source: Preliminary Estimates from EPA Stage I NESHAP
project on gasoline marketing.
3-5
-------
3.2 EMISSION SOURCES
Emission sources described in this section are divided
into service station Stage I emissions (gasoline transfers
into the station underground storage tanks) and service
station Stage II emissions (automobile refueling emissions).
3.2.1 Service Station Stage I Emissions
Gasoline vapor or volatile organic compound (VOC)
emissions occur when gasoline being delivered to the service
station displaces vapors to the atmosphere (as described
earlier). Under a typical gasoline delivery, a hose is
connected from the delivery truck to a ground level fitting
that is attached to the underground gasoline storage tank
(see Fi-gure 3-1) . The gasoline is allowed to drop from the
delivery truck into the underground tank. This activity is
often called "the service station drop" or "dropping a load
of product". Displaced vapors are emitted to the atmosphere
through the underground tank vent. Submerged loading,
consisting of a tube installed to within 6 inches of the
bottom of the tank, significantly reduces emissions because
turbulence caused by the splashing of the delivery product
in the underground tank is minimized.
When Stage I emission controls are used, displaced
vapors are collected and routed back into the delivery truck
using a combination of pipes and hoses (see Figure 3-2).
Stage I emissions from service stations and the resulting
technology are not the subject of this report but have been
included in the discussion for completeness. These
emissions have been the subject of several EPA programs and
further information can be obtained in other EPA
publications.3-4'5'6'7 While tank truck unloading (Stage I)
and vehicle refueling (Stage II) are separate events,
defective Stage I equipment (leaking seals, missing caps,
etc.) can adversely affect the efficiency of a Stage II
systenu
3-6
-------
Underground Storage Tank
Storage Tank Vent Pipe
Loading of Service Station Underground Storage Tank
(A) With No Controls.
Mir
Meter-"
Underground Storage Tank
Storage Tank Vent Pipe
Service Station Vehicle Refueling With No Controls
Figure 3-1. Uncontrolled Service Station operations
3-7
-------
Vapor Manifold Piping
Underground Storage Tank
Storage Tank Vent Pipe
Loading of Service Station Underground Storage Tank
(A) With Vapor Balance System (Stage I Controls),
Coaxial Vapor/Liquid Hose
Meter
Vapor Recovery
Nozzle
1 8,
g Underground Storage Tank
Storage Tank Vent Pipe
Service Station Vehicle Refueling With Vapor Balance
(B) System (Stage n Controls).
Figure 3-2.
Controlled Service Station Operations
(Stage I and Stage II)
3-8
-------
3.2.2. Vehicle Refueling Emissions
3.2.2.1 Vehicle Refueling. Gasoline vapor/VOC
emissions occur when liquid from the underground tank is
dispensed into the vehicle fuel tank. Vapors contained in
the fuel tank are displaced back through the vehicle
fillneck and are emitted to the atmosphere (see Figure 3-1).
With the installation of Stage II vapor recovery equipment,
displaced vapors are captured at the vehicle fillneck and
routed back to the underground tank. Figure 3-2 illustrates
the basic Stage II vapor recovery concept. Detailed
descriptions of the Stage II vapor recovery equipment and
discussions of emission reductions can be found in Chapter
4. Factors influencing emissions and estimates of emissions
are presented later in this chapter.
3.2.2.2. Spillage. VOC emissions from the vehicle
refueling operation can also occur when loading the vehicle
at a rate faster than the displaced vapors can be released.
When this occurs liquid is forced up the fillneck and can
cause "spitback" of liquid back out of the vehicle fillneck.
Overfilling of the vehicle can also cause liquid spillage.
Overfills can occur due to a failure in the nozzle shutoff
mechanism or can occur due to operator error (repeated
"topping off" of the vehicle tank). Small amounts of liquid
drips can also be spilled due to wetted nozzle tips upon
removal from the vehicle and vapor condensation on cool
nozzle surfaces.
3.2.2.3. Breathing/Emptying Losses. Emptying losses
occur when gasoline is pumped out of the service station
underground tank to refuel a customer's automobile fuel
tank. Air is drawn into the underground tank, through the
underground tank vent pipe, to replace the volume of liquid
removed. Prior to any gasoline being removed from the tank,
the liquid and vapors in the underground tank are at
equilibrium and the vapor space above the liquid is
essentially saturated. When liquid is pumped from the tank
and air is drawn in through the vent, the vapor space above
3-9
-------
the liquid is no longer in equilibrium with the liquid. A
small amount of liquid evaporation takes place in an attempt
to again saturate the vapor space above the liquid. This
evaporation causes an increase in volume in the vapor space
and this excess volume is pushed out the underground tank
vent pipe. The portion of vapors pushed out the vent is
called the emptying loss.
Stage II vapor recovery equipment helps to controls
this emptying loss by returning essentially saturated vapors
from the vehicle fuel tank back to the service station
underground tank to replace the liquid removed. Because the
return vapors are saturated and equal in volume to the
liquid removed, equilibrium in the tank is maintained,
product evaporation does not take place, and emptying loss
emissions do not occur.
Breathing loses in fixed volume storage tanks are
caused by vapor and liquid expansion and contraction due to
diurnal temperature changes. As temperatures increase,
vapor volume increases pushing vapor out of the vent pipe
(out-breathing). When temperatures decrease, vapor volume
decreases and air is drawn into the tank (in-breathing),
Breathing loss emissions are minimal at service stations
since storage tanks are located underground, insulated by
the earth, and have a very stable temperature profile.
However, breathing losses from service station storage tanks
are becoming more prevalent due to the popularity of above
ground storage tanks and the installation of vaulted
underground storage tanks. Above ground storage tanks are
more susceptible to temperature and pressure changes and
thus are more likely to experience both vapor growth and
vapor shrinkage. It is also reported that the double wall,
or "vaulted" underground storage tanks that are being
installed to comply with underground storage tank (UST)
regulations are more susceptible to thermal effect and
therefore breathing losses.8'9
3-10
-------
3.3 FACTORS INFLUENCING EMISSIONS
Many studies have been done to evaluate the factors
that affect refueling emissions. A recent study by EPA's
Office of Mobile sources (OMS) empirically derived an
equation that predicts the emissions from an automobile
refueling event.10 This testing consisted of controlled
vehicle refueling inside a shed with sensors to gather fuel
tank temperature, liquid dispensed temperature, and
displaced vapor. Emissions testing was conducted on a
variety of light-duty vehicles, with varying fillneck
configurations, and on light-duty trucks. The following
sections describe the different factors that influence this
emission factor equation.
3.3.1 Reid Vaoor Pressure (RVP)
Certainly one of the most important factors affecting
the emissions from automobile refueling is the volatility of
the gasoline. A less volatile gasoline will create less
emissions when transferred than a more volatile gasoline.
Reid vapor pressure (RVP) is a common measure of fuel
volatility and represents the vapor pressure of the fuel at
100"F. RVP is a standard industry measure of fuel
volatility. Although RVP is a measure of fuel volatility at
100°F, the empirical emissions equation described below
(3.4.1) adjusts this volatility to reflect actual
temperature conditions.
The RVP of gasoline is adjusted through blending at the
refinery to account for temperature and pressure different-
iations across the country. In the summer when warm
temperatures enhance volatilization, gasolines can be
blended with a lower RVP and still provide ample vaporiza-
tion for combustion in the vehicle engine. Reducing RVP in
the summer, therefore, reduces emissions from gasoline
transfers without reducing vehicle performance. Too high an
RVP in the summer can create excess volatilization in the
engine causing vapor lock. During the winter months when
cold temperatures inhibit volatilization, gasolines can be
3-11
-------
blended with a higher RVP to ensure sufficient volatiliza-
tion for engine start-up and operations. This increase in
RVP when temperatures decrease and decrease in RVP when
temperatures increase is an attempt to provide a uniform
fuel volatility for smooth engine performance all year.
Information on winter/summer actual RVP samples are
taken throughout the year in selected areas. This
information is compiled and published by the National
Institute for Petroleum and Energy Research (NIPER)
organization. This data is based on fuel surveys and fuel
analyses conducted throughout the country.11
Fuel RVPs can be blended to adjust for certain altitude
and temperature variations in specific geographical areas.
On June 11, 1990, EPA promulgated limits for RVP in the
summer for all States.12 These limits will reduce fuel RVP
to 9.0 or below in most States in the summer months.
However, the RVP requirements proposed in the May 29, 1991,
Federal Register13 indicate that RVPs less than 9.0 will
only be required during the summer months in ozone
nonattainment areas. The remaining areas in States with
lower RVP limits need only meet 9.0. Table 3-3 summarizes
the RVP restrictions by month for each State for the entire
year.1*'15 The weighted averages presented are weighted by
the monthly fuel consumption presented in Table 2-1. In
addition, the summer weighted average RVP is calculated
using the values in the table (i.e. values less than 9.0
RVP) and is therefore representative of nonattainment areas
for those States. Attainment area RVP would be higher since
summer RVP is not regulated below 9.0. For those States
where an RVP restriction less than 9.0 appears in the summer
months, this more stringent restriction applies only to
3-12
-------
TABLE 3-3.
1992 AND BEYOND RVP LIMITS BY MONTH
AND BY GEOGRAPHIC LOCATION
FEI
HA*
•eld V«por Prwture (ptl)
HM JUN JUL AUC
SEP
OCI
tnv
Wel»httd Avtrag*
Ulntw Annual
DEC (Apr-S*p) (Oet-Ntr)
ALABAMA
ALASKA
ARIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
OEUWAIE
OIST. OF COL.
FLORIDA
HAHAII
IDAHO
ILLINOIS
INDIANA
10UA
KENTUCKY
LOUISIANA
MAINE
MARYLAND
NICNI6AN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA
HEW HAMPSHIRE
HEW JERSEY
HEW MEXICO
NEW YORK
13.5
15.0
13.5
14.2
13.6
15.0
15.0
15.
15.
13.
.
11.
»fL
.0
15.0
15.0
15.0
»n
*w
15.0
1S.5
15.0
15.0
Wn
• V
15.0
15.0
13.5
1S.O
15.0
15.0
14.2
15.0
15.0
HA
.T
15.0
13.5
15.0
12.5
13.5
13.4
14.2
15.0
15.0
14.2
13.5
Hm
»5
11.5
M*
•Z
15.0
15.0
15.0
W9
»2
14.2
13.5
15.0
15.0
»A
• V
15.0
15.0
13.5
14.2
15.0
15.0
13.4
15.0
15.0
»9
•2
15.0
12.5
15.0
10.8
12.5
12.6
12.5
14.2
14.2
13.5
12.5
WB>
•3
11.5
«•
.5
14.2
14.2
14.2
«•
• 5
13.5
12.5
14.2
14.2
H«
.2
14.2
14.2
12.5
13.
It.
14.
12.
14.
14.
««
11*
14.
11.5 9.0
15.0 14.2
10.0 ,0
11,5 .0
11.* .0
11.5 .0
1J.S .0
1J.5 .0
12.5 .0
11.5 .0
11.5 1 .5
II.O .0
13.5 .0
12,5 .0
12.5 .0
11.5 .0
1J.5 .0
13.5 .0
13.5 .0
13.5 .0
11.5 .0
12.5 .0
12.5 .0
12.5 .0
11.2 .0
13.5 .0
13.5 .0
13.5 .0
7.8
13.5
7.8
7.8
7.8
7.8
9.0
9.0
7.8
7.8
11.5
9.0
9.0
9.0
9.0
7.8
9.0
7.8
9.0
9.0
7.8
7.8
9.0
9.0
7.8
9.0
9.0
9.0
7.
13.
7.
7.
7.
7.
9.0
9.0
7.8
7.8
11.5
9.0
9.0
9.0
9.0
7.8
9.0
7.8
9.0
9.0
7.8
7.8
9.0
9.0
7.8
9.0
9.0
9.0
7.
13.
7.
7.
7.
7.8
9.0
9.0
7.8
7.8
11.5
9.0
9.0
9.0
9.0
7.8
9.0
7.8
9.0
9.0
7.8
7.8
9.0
9.0
7.8
9.0
9.0
9.0
7.8
14.2
7.8
7.8
7.8
7.8
9.0
9.0
7.8
7.8
11.5
9 A
.8
9.0
9.0
9.0
9.0
7.8
9.0
7.8
9.0
9.0
7.8
7.8
9.0
9.0
7.8
9.0
9.0
9.0
11.5
15.0
9.5
12.5
10.5
10.8
11.5
12.5
12.5
11.5
«e
•3
11.5
Wn
.B
12.5
12.5
12.5
Wft
• 0
12.5
11.5
13.5
12.5
We
.5
13.5
12.5
11.5
12.5
12.5
10.8
10.2
13.5
13.5
Wn
*B
13.5
12.5 13.5 8.*
15.0 15.0 13.9
10. 12.5 8.4
13. 14.2 .5
12. 13.6 .6
12. 14.2 .6
14. 15.0 .7
14. 15.0 .7
14. 15.0 .8
12. 13. .7
H«*
. u* «
11. 11. 11.
Wtt a
* !*• ™»
11. 14. 9.
14. 15.0 9.
14. 15.0 9.
U4 j 9 jBt
, 14 *< 9«
14. 15.0 .6
12. 1S.S .*
14. 15.0 .6
14. 15.0 .0
M«B A 9
. 1>«V «r
14.2 15.0 .7
14.2 15.0 .7
12.5 13.5 .*
13.5 14.2 .7
14.2 15.0 .5
12.5 14.2 .5
11.6 13.4 .5
14.2 15.0 .7
14.2 15.0 .7
«C 1* C •
.9 \9*y > ** Mt
1%* 1Z*D
14. 12.0
14. 11.8
12. 10.7
13. 11.1
14. 11.7
13. 11.4
12. 10.4
14. 12.0
14. 12.1
K«D *
. 10* J
14.5 12.0
-------
TABLE 3-3.
1992 AND BEYOND RVP LIMITS BY MONTH AND
GEOGRAPHIC LOCATION (CONTINUED)
ffi
(eld Vapor Prnturt (psl)
AM KAt MM JW. AU6
SEP
OCT
NOV
DEC
Utlghtvd Av*r*«t
Ulnttr Anrtttl
(Apr-S«p) COct-Nwr}
NORTH CAROLINA 14.2 13.! 13.5 12.S 1
NORTH DAKOTA 1S.O 1S.O 14.2 13.5
OHIO 1S.O 1S.O 14.2 13. S
fWl AUfMlA 11. 9 tV * 19 C 11 C
OKLAHOMA 14*2 13 15 lc*> 11.5
ju.»juuj «K A fi 9 11 C IV C
ORcvOll 15.0 14.2 13.9 13.3
PENNSYLVANIA 15-0 15.0 14.2 13.5
INODf ISLAM 15. 0 15.0 14.2 13.S
SOUTH CAROLINA 13. S 13. S 13.5 12.5
w SOUTH DAKOTA 15.0 11.0 14.2 12.5
1 TENNESSEE 14.2 13.S 11.5 12.5
£ TEXAS 13.5 13.0 11.6 10.8
•IV*M 1C A 1JL 9 fV C 19 C
UTAH 15.0 14*2 13*5 12.5
venom is.o 15.0 u.2 is.s
VIRGINIA 15.0 14.2 11.5 12.
WASHINGTON 15.0 15.0 14.2 13.
WIT VIRGINIA 15.0 15.0 14.2 13. 1
WISCONSIN 15.0 15.0 14.2 13. 5
VFTMIW 19.0 11.0 14.2 12. «
>.0 7.8
.0 .0
.0 .0
.0 .0
.0 .0
.0 9.0
.0 7.8
,0 7.8
.0 9.0
.0 7.8
.0 9.0
>.0 9.0
».0 9.0
>.0 9.0
!••••««•••«*«•
7.8 7.8 7.8 12.
9.0 9.0 9.0 12.
9.0 9.0 9.0 12.
9.0 9.0 9.0 13.
9.0 9,0 9.0 IS.
9.0 9.0 9.0 10.
7.8 7.8 7.8 12.
7.8 7.8 7.8 10.
.0 9.0 9.0 13.
.8 7.8 7.8 12.
.0 9.0 9.0 12.
.0 9.0 9.0 12.
.0 9.0 9.0 12.
.0 9.0 9.0 10.
5 13. 14.2
5 14. 15.0
14. 15.0
H*jt 9
» 14.2
H14 A
. 14. 0
U.2 15.0
14.2 IS.O
»« 1* «
.5 13.5
12.5 14.2
13.5 14.2
12.5 1S.S
»K €« %
.3 14«2
14.2 15.0
14.2 15.0
14.2 15.0
14.2 15.0
14.2 15.0
12.5 14.2
.8 13.1
.7 14.
.7 14..
«49 i
1«.
0»
13.
.r u.
.7 14.
Q*«
13.
.5 11.
.8 13.
.5 12.
7*W
13.
.6 14.
.8 14.
.7 14.
.7 14.
.7 14.
.5 13. <
i 11.1
I 11.7
I 11.9
>1A f
lw*r
• « 9
11*2
12.0
12.1
*« A
11. V
11.3
11.1
10.4
Wn
*T
12.0
11.3
11.9
11.9
11.9
t 11.5
Source : fix eoMunlcatfan
•nd JIM 11, 1990
fro* lob Johnion, tPA/OMS, April 10, 1991.
•nd Nay 29, 1991 FEDERAL REGISTERS
Nationwide Anruil Avertgt: 9.4
Hofwttilnwnt Amucl Avtragts 9.2
11.4
11.3
-------
nonattainment areas within the State. RVP in non summer
months is typically blended to conform to limits suggested
by ASTM and is not usually regulated by EPA.
3.3.2 Liquid Temperature
Along with fuel volatility, the temperature of the fuel
being dispensed and the temperature of the vehicle fuel tank
affect the rate in which emissions occur. The warmer the
temperature of the dispensed liquid or the vehicle fuel tank
the more volatile the liquid becomes and the more emissions
occur. Also, the temperature difference between the
dispensed liquid and the liquid in the fuel tank can affect
emissions. The loading of cool dispensed fuel into a warm
tank will decrease emissions (like vapor shrinkage) and the
loading of warm fuel into a cold vehicle tank can increase
emissions (like vapor growth). The more typical situation
is where you have cool liquid being dispensed into a warm
vehicle tank. The empirically derived emission factor
equation accounts for these temperature differences.
As with RVP, these key temperature parameters will vary
with time of year and with geographical location. Table 3-4
presents dispensed fuel temperature presented by month for
several regions in the country (Figure 3-3 indicates the
regional boundaries).16 As would be expected, dispensed
fuel temperatures increase in the summer when RVPs decrease.
Table 3-5 presents average annual fuel differentials
between the dispensed fuel and the fuel in the vehicle tank.
Data are presented by region for an annual average AT, plus
values for summer and winter months.17 In addition, data
are presented for a 5-month (May-September) and 2-month
(July and August) ozone season.
3.4 EMISSION FACTOR CALCULATIONS
3.4.1 Vehicle Refueling
As discussed in Section 3.3, EPA Office of Mobile
Sources empirically derived an equation to estimate
3-15
-------
TABLE 3-4
MONTHLY AVERAGE DISPENSED LIQUID TEMPERATURE
Dispensed liquid Temperature (degrees F) Weighted Average
Sinner Winter Annual
MN fit MAR APR HAT JUN JUL AUG SEP OCT MOV DEC (Apr-Sep) (Oct-Nar)
U
1
o\
Natlmwl Average 51 54 54 58 69 76 82 at 76 70 62 Si
legion 1 43 45 48 51 66 74 78 78 72 66 59 46
Region 2 69 ?4 73 80 84 87 90 91 78 85 83 73
Region 3 54 57 61 67 76 8Z 82 84 79 76 67 54
legion 4 50 51 41 47 63 74 88 85 83 75 63 52
Region S S* - - - n 77 83 83 79 74 67 58
Region 6 - 48 49 S3 59 63 73 71 60 49 42
74 58 66
70 SI 61
85 76 81
79 62 70
74 56 65
79 63 72
64 SO 57
Source : NeJtnalty, Hlcftael Mid Dlckenaan, J.C. Smiury and Aiwlysll of Dili Frooi Gasoline Tenpernture Survey
Conducted ly A**rle«n Ntroteut Institute. Kid!en Corporation, Hay 1976.
Regional boundaries defined In figure S.S.
-------
I
H
•4
stations
Figure 3-3
Boundaries
-------
TABLE 3-5. SEASONAL VARIATION FOR TEMPERATURE DIFFERENCE
1ABW, -J a. BETWEEN DISPENSED FUEL AND VEHICLE FUEL TANK (AT), F
1 National Average
00
Region 1
Region 2
Region 3
Region 4
Region 5
mmmmmmmmmmmmmmmmmmm
Average
Annual
4.4
5.7
4.0
3.7
5.5
0.1
Temperature
Sunnier
5-Nonth
Ozone Season
(May-Sep)
9.4
11.5
7.5
7.1
12.1
5.1
2 -Month
Ozone Season
(Jul-Aug)
9.9
12.5
8.2
7.0
13.3
3.2
Source : Rotlman, David, and Johnson, Robert. Technical Report - Refueling Emissions From Uncontrolled Vehicles.
EPA/QMS, EPA-AA-SOSB-85-6, June 1985.
-------
refueling emissions based on test data. This equation is as
follows:
I. '
where:
RVP
AT
264.2[ (-5.909) - 0.0949(AT) + 0.0884(T0)
-I- 0.485 (RVP)J
Emission rate, milligrams of VOC per
liter of liquid loaded
Reid vapor pressure, psia
Difference between the temperature of
the fuel in the automobile tank and the
temperature of the dispensed fuel, *F
TD = Dispensed fuel temperature, °F
Using this emission factor equation, and the RVP and
temperatures found in Tables 3-3, 3-4, and 3-5, automobile
refueling emission factors can be derived for specific
geographic locations and for different seasons of the year.
Emission factors calculated using this equation should allow
the estimation of emissions from automobile refueling for
any area of the country. This approach is certainly more
accurate than using the single value provided in EPA's
Compilation of Emission Factors (AP-42).18
Table 3-6 illustrates how these emission factors can
vary from location to location and by time of year for each
State. Using the emission factor equations indicates
variations of over 40 percent between summertime emissions
rates found in Colorado (1,080 mg/L) and Florida (1,550
ing/liter) . This indicates that an error would be introduced
in emission planning activities if a single factor were
used.
While this methodology has been used in prior EPA
studies19*20 to estimate refueling emissions, it should be
noted that revised State implementation plan (SIP) emission
inventory guidance issued by EPA in 199121 recommends that
refueling emissions be calculated using emission factors
3-19
-------
TABLE 3-6. MONTHLY AND GEOGRAPHIC VARIATIONS IN REFUELING EMISSION FACTORS
ALABAMA
ALASKA (a)
ARIZONA
ARKANSAS
CALIFORNIA (b)
COLORADO
CONNECTICUT
DELAWARE
DIST. OF COL.
FLORIDA
co
^ GEORGIA
0 HAWAII (a)
IDAHO Ca.bl
ILLINOIS
INDIANA
IOWA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA (b)
NEW HAMPSHIRE
JAN
1760
1570
1440
1850
1550
1590
1370
1370
1370
1760
1760
1120
1540
1370
1370
1590
1590
1370
1760
1370
1370
1370
1370
1590
1760
1590
1590
1590
1630
1370
FEB
1870
1640
1380
1870
1750
1510
1420
1420
1320
1870
1870
1190
1400
1420
1420
1610
1510
1320
1870
1420
1420
1420
1420
1610
1870
1510
1610
1610
1750
1420
MAR
1720
1640
1260
1720
1850
1060
1390
1390
1300
1720
1720
1190
1330
1390
1390
1280
1060
1300
1720
1390
1390
1390
1390
1280
1720
1190
1280
1280
1800
1390
APR
1610
1490
1090
1610
1670
720
1140
1140
1010
1610
1610
1050
1060
1070
1140
840
720
1010
1610
1140
1140
1140
1140
970
1610
840
840
840
1620
1140
MAY
1380
1650
1020
1380
1060
770
970
970
980
1380
1380
1300
750
920
970
770
770
870
1380
970
970
970
970
870
1380
770
770
770
1100
970
JUN
1450
1720
1060
1450
1140
870
1050
1050
1050
1450
1450
1470
840
1050
1050
1030
1030
1050
1450
1050
1050
1050
1050
1030
1450
1030
1030
1030
1140
1050
JUL
1460
1860
1080
1370
1280
1150
1150
1150
1150
1520
1460
1610
1060
1120
1150
1350
1200
1150
1460
1150
1150
1150
1150
1350
1460
1200
1350
1350
1280
1150
AUG
1390
1840
1110
1390
1280
1080
1150
1150
1150
1550
1390
1580
1080
1120
1150
1280
1130
1150
1390
1150
1150
1150
1150
1280
1390
1130
1280
1280
1280
1150
SEP
1240
1810
990
1240
1190
1080
1110
1010
1010
1240
1240
1470
1030
1010
1010
1240
1240
1010
1240
1110
1010
1110
1110
1240
1240
1240
1240
1240
1190
1110
OCT
1880
2020
1440
2000
1620
1630
1720
1590
1590
1880
1880
1570
1240
1590
1590
1850
1630
1590
1880
1720
1590
1720
1720
1850
1880
1850
1850
1630
1580
1720
NOV
1960
1830
1400
2080
1660
1570
1640
1640
1640
1960
1960
1380
1216
1610
1640
1790
1570
1640
1960
1640
1640
1640
1640
1790
1960
1700
1790
1570
1600
1640
Weighted Average
DEC Sumner Winter Annual
(Apr-Sep) (Oct-Mar)
1850
1640
1310
1940
1650
1530
1440
1440
1440
1850
1850
1190
1260
1390
1440
1640
1530
1440
1850
1440
1440
1440
1440
1640
1850
1530
1640
1530
1620
1440
1420
1730
1060
1400
1270
950
1090
1070
1060
1460
1420
1420
970
1050
1080
1090
1010
1040
1420
1100
1080
1090
1090
1130
1420
1030
1100
1090
1270
1100
1840
1740
1370
1910
1680
1460
1500
1480
1440
1840
1840
1280
1320
1470
1480
1640
1480
1450
1840
1500
1480
1500
1500
1630
1840
1560
1630
1530
1660
1500
1630
1740
1220
1630
1470
1200
1290
1260
1250
1650
1630
1350
1150
1260
1270
1350
1210
1230
1620
1290
1280
1290
1290
1360
1630
1290
1340
1300
1460
1290
-------
TABLE 3-6. MONTHLY AND GEOGRAPHIC VARIATIONS IN REFUELING EMISSION FACTORS (CONTINUED)
OJ
I
NEW JERSEY
NEW MEXICO
NEW YORK
NORTH CAROLINA
NORTH DAKOTA
OHIO
OKLAHOMA
OREGON (»,b)
PENNSYLVANIA
RHODE ISLAND
SOUTH CAROLINA
SOUTH DAKOTA
TENNESSEE
TEXAS
UTAH
-------
generated by MOBILE 4.1, EPA's mobile source emission factor
computer model. MOBILE4.1 utilizes the sane equation
presented above to calculate a refueling emission factor.
User supplied inputs for temperature and RVP are used to
calculate an emission factor based on gasoline throughput
(gm/gal) . MOBIL14.1 also will convert this emission factor
to one based on VMT by using assumptions for^ the on-road
automobile population and the fuel economy for each model
year. Hiere is uncertainty introduced by using VMT as the
parameter for calculating refueling emissions. First, the
fact that a vehicle travels through a certain area does not
indicate that the vehicle is refueled in the same area, and
second, the use of fuel economy introduces another layer of
uncertainty to the calculation. In the absence of accurate
throughput data, refueling emissions may be estimated using
VMT. However, it is suggested in MOBIL14.1 guidance that
refueling emissions be calculated using throughput data
instead of VMT.22
3.4.2 Spillage
Several recent studies have been conducted comparing
the occurrences of spillage during refueling events both
with and without Stage II vapor recovery equipment. The
studies are: (1) a 1989 study by the American Petroleum
Institute23; (2) a 1990 study by the California Air
Resources Board24,' and (3) a 1983 study by the Bay Area Air
Quality Management District.25 A fourth study was conducted
in 1987 by Lundberg.26 The Lundberg study provided some
simplified frequency information but no quantification of
spillage or emissions. The survey contained only
observances of spillage along with other questions and
observations taken during refueling episodes. Since no
quantification of spills was contained in the study, it is
not summarized here.
The three studies were similar in that they observed
refueling at both conventional and Stage II systems,
documented spillage frequency, and estimated the quantity of
3-22
-------
spillage that occurred. Spillage quantities were estimated
by correlating spill area measured on the ground with volume
quantities of liquid gasoline spilled. Table 3-7 summarizes
the results of these studies.
The API study was conducted at 20 "well maintained"
Stage II systems in the Washington, DC area and 20
conventional systems in Baltimore. Considerable effort was
taken to assure that the Stage II and conventional stations
were comparable in throughput, number of nozzles, and
location (urban inner city). Spills were quantified by
measuring wetted surface area caused by the drip or spill
that occurred during the refueling cycle.
Inspectors/observers were trained by spilling specific
liquid quantities and measuring the resulting spill area.
Spill areas were calibrated at each test site to take into
account differences in surface porosity, fuel character-
istics, and ambient conditions. The API study found an
increase in spill frequency with Stage II equipment and an
increase in spill quantity.
The CARS study was similar to the API study in
methodology using spill size versus quantity techniques. In
addition to measurable spills on the ground, GARB included
spills along the side of the vehicle. The CARB study took
place at 31 Stage II systems in Sacramento and 21
conventional stations north of Sacramento. Data were
reported for all spills and adjusted to account for one
large spill that CARB felt biased the results. API made no
adjustments to the data collected at the Washington, DC and
Baltimore stations for any large spills. To convert spill
siEe/volume data in the CARB study to quantity data, two
assumptions had to be used: (1) gasoline density was 0.67
gm/ml (the same used in the Stage II recovery credit
calculations), and (2) the average volume per refueling
event was 10 gallons. The CARB study found a lower
3-23
-------
TAIll 3-7
SUMMARY OF STAGE 11/CONVENTIONAL REFUELING SPILLAGE DATA
OJ
I
ro
Bay Area ("1983)
Reported Conven.
Data
Balance System
Post '78 Balance
Vacuum Assist
Post '78 Vac. Ass.
Red Jacket
Post '78 Red Jacket
CARB Study (July 1991)
API Study (June 1919)
Observations
Conven, Stage II
6,750
1,254
310
737
118
83
9
1,496 1,515
1,357 1,271
Frequency
Conven. Stage II
0.32
0.39
0.40
0.31
0.28
0.13
0.00
0.30 0.22
0.63 0.66
gm/liter
Conven. Stage II
3.51
1.15
0.43
0.66
0.32
0.67
0.00
2.21 * 1.59
1.20 1.74
Pv*oallon ng/lfter
Conven. Stage II Conven Stage II
0.30 80.0
0.12
0.05
0.07
0.03
0.08
0.00
0.22 ** 0.16 58.3
0.14 0.22 36.9
31.6
13.9
17.6
8.S
19.6
0.0
41.9
58.9
* Assumed gasoline density of .67 gm/«l.
** Assumed 10 gallons per refill event.
-------
frequency of spills and smaller quantities of spills with Stage
II equipment. It should be noted that spillage determinations
are part of the certification procedures for Stage II equipment
in California. To pass certification, the Stage II equipment
must have spillage quantities less than conventional equipment.
The third study was conducted by the Bay Area AQMD. The
results of this study was obtained from the Bay Area, but no
narrative was supplied. From the data supplied and a
conversation with Bay Area AQMD it was determined that the test
program was similar to that of the CARB and API studies. The
conventional nozzle study dates back to a 1974 study by Scott
Environmental. This conventional nozzle study by Scott was the
basis for the AP-42 emission factor for spillage from automobile
refueling (80 mg/liter). The Stage II data were obtained from
facilities in the Bay Area. The Bay Area data indicated a slight
increase in spill frequency with Stage II equipment but a
significantly lower emission rate.
It is difficult to draw any specific conclusions on the
relative merit of the studies. Each appeared to incorporate
similar procedures, however, slightly different results were
obtained. The results of all studies are in the same order of
magnitude and in the same approximate range. This further
complicates the task of evaluating spillage information. It is
impossible, based on this data, to conclude one way or the other
on whether Stage II or conventional refueling results in higher
spillage. This difficulty in concluding a definitive spillage
quantity must be put in perspective. The difference in this
spillage data represents less than one percent of the emissions
from the total refueling event.
3.4.3 Emptying Losses
Emissions have also been reported at service stations due to
storage tank emptying and breathing losses. Breathing losses are
attributable to gasoline evaporation due to barometric pressure
and temperature changes. Breathing loses in fixed volume storage
tanks are caused by vapor and liquid expansion and contraction
due to diurnal temperature changes. As temperatures increase,
3-25
-------
vapor volume increases pushing vapor out of the vent pipe (out-
breathing) . When temperatures decrease, vapor volume decreases
and air is drawn into the tank (in-breathing). Breathing loss
emissions have traditionally been minimal at service stations
since storage tanks have generally been located underground,
insulated by the earth, with a very stable temperature profile.
However, breathing losses from service station storage tanks are
becoming more prevalent due to the popularity of aboveground
storage tanks and the installation of vaulted underground storage
tanks. Aboveground storage tanks are more susceptible to
temperature and pressure changes and thus are more likely to
experience both vapor growth and vapor shrinkage. It is also
reported that the double wall, or "vaulted" underground storage
tanks being installed to comply with underground storage tank
(UST) regulations are more susceptible to thermal effect and
therefore breathing losses.27'28
Emptying losses occur when gasoline is withdrawn from the
tank allowing fresh air to enter. This enhances evaporation
(i.e., vapor growth) and causes vapors to be vented from the pipe
as the saturated gasoline vapors tend to occupy a larger volume
than air. EPA's AP-42 cites an average breathing emission rate
of 120 milligrams per liter of throughput.
This original source for this factor was a Journal of the
Air Pollution Control Association November 1963 article based on
a study by the Air Pollution Control District of Los Angeles
County (LAAPCD). This article was entitled "Emissions from
Underground Gasoline Storage Tanks", and lists as authors Robert
Chass, Raymond Holmes, Albert Fudurich, and Ralph Burlin of the
Los Angeles District.29 This article describes emptying losses
as follows.
When an automobile is fueled, gasoline is pumped
from the underground tank, causing air to be inhaled
through the vent pipe, the volume being approximately
equal to the volume of gasoline withdrawn. The air
then becomes saturated with gasoline vapors, tending to
occupy a larger volume. This in turn, causes the
vapor-air mixture to exhaust from the underground tank
until a pressure equilibrium is attained.
3-26
-------
The mg/1 emission factor listed in AP-42 was estimated in
this study by measuring air expelled from the vent pipe after
vehicle fueling and applying a theoretical gasoline vapor to air
ratio of 40 percent. They concluded that it was impractical, in
their study, to collect representative vapor samples for
analysis. While the emission factor of one pound per thousand
gallon of throughput (approximately 120 mg/1) was presented in
this study, it also discussed complexities with estimating these
emissions. The study concluded:
Factors affecting the breathing losses are complex
and interrelated, depending on the service station
operation, pumping rate, frequency of pumping, ratio of
liquid surface to vapor volume, diffusion and mixing of
air and gasoline vapors, vapor pressure and temperature
of the gasoline, the volume and configuration of the
tank, and the size and length of the vent pipe.
Because of these many variables involved, much more
data from a number of representative retail stations
would be necessary before an accurate determination of
overall, basin-wide breathing losses could be made.
Since the time of this original analysis, several studies
have been conducted to attempt to account for many of these
variables. These range from studies that conclude there are no
VOC emptying losses to those reporting emissions much higher than
those predicted by the AP-42 emission factor.
Dr. R.A. Nichols has studied this subject extensively
throughout the 1970s and 1980s. In a 1987 paper on the
subject30, the conclusion is that the model used in the LAAPCD
analysis ignored the effect of the vent line. Dr. Nichols
states:
Air enters a nearly underground tank containing
saturated vapor. Air will spread over a large and
heavier vapor layer enhancing diffusion into this
layer. As the surface layer gains vapor, the lighter
upper vapor, which is essentially air, is vented from
the tank through the vent line. The air-vapor mixture
expelled from the tank to the vent line occupy only a
small fraction of the vent line volume. The air-vapor
mixture remains in the vent pipe for some time because
of low diffusion rate. Subsequently, this mixture is
inhaled back into the tank in the next refueling.
Consequently, the vent line acts as a buffer to
3-27
-------
effectively ensure that only air enters and leaves the
vent during intermittent refueling.
Dr. Nichols indicates that vapor emissions could only occur
during periods of long refueling inactivity. He concludes that
high fueling activity followed by long periods of inactivity will
lead to the highest (and possibly the only) vapor venting
emissions. This paper did not provide any emission factor for
these emissions.
The California Air Resources Board (GARB) conducted a study
to estimate storage tank breathing losses in 1987.31 Emissions
were measured at a low throughput (15,000 gallons per month per
tank) station and a high throughput (50,000 gallons per month per
tank) station. The study found different results for the two
stations. The emission factor calculated for the low throughput
station was 0.92 Ibs VOC per 1000 gallon throughput (110 mg/1),
and 0.21 pounds per 1000 gallon (25 mg/1) for the high throughput
station. Observations made during the testing indicated that
mass emissions from the underground storage tanks appeared to
occur during periods when dispensing of product was the lowest,
that emissions were at a minimum during conditions of near
continuous fuelings, and that the highest mass emissions occurred
during intermittent vehicle fuelings followed by relatively long
periods of dispensing inactivity. The differences in emission
factors at the high and low throughput stations are explained in
these observations.
The National Institute for Petroleum and Energy Research
(NIPER) conducted a study and reached conclusions partially in
agreement with those of both Dr. Nichols and GARB.32 NIPER1 s
study concluded that no vent losses would occur if the dispensing
frequency were high enough and that vent losses would be markedly
reduced if the height of the vent was increased. The rationale
for the origin of emissions agreed with the discussion provided
in the original LAAPCD study. This was that emissions were due
to l) air induction through the vent, 2) dilution of the
hydrocarbon vapor in the tank, 3) saturation of the diluted vapor
by evaporation of the liquid fuel, resulting in increased
3-28
-------
pressure in the tank. If this pressure was greater than that
exerted by the column of vapor in the vent, emissions resulted.
The emissions measured for a high flow stations were 0.85 and
1.05 grams per gallons dispensed (225 and 277 mg/1,
respectively).
A comparison of the CARB and NIPER studies shows that the
NIPER emission factors are much higher than those from CARB.
Recognizing this discrepancy, CARB and NIPER met on August 21,
1987 to discuss the differences.33 The conclusion reached at
this meeting was that NIPER's results should be adjusted because
the dispensing period (8 hours) during NIPER"s tests was not
considered representative of the effective dispensing period at a
high volume station. Adjustments were made and it was determined
that a more appropriate emission factor for the NIPER data is 0.6
lbs/1000 gallons (72 mg/1) for a high throughput station.
In summary, these studies indicate that the emissions from
storage tank emptying are affected by several factors, most
notably the height of the vent pipe and the vehicle fueling
activity. For the purposes of the analysis in this document, it
is believed that the AP-42 factor of 120 mg/1 represents an
emission factor that may be very conservative, but is not
unrealistic.
3.5 MODEL PLANT EMISSION ESTIMATES
Model plants, as described in Chapter 2, are used to
represent the industry for cost and emission estimation purposes.
The data presented earlier in this chapter and in Chapter 2 were
used to calculate emissions for each model plant. Table 3-8
summarizes model plant emissions using an emission factor
calculated with the overall national annual average RVP of 11.4
psi, a AT of 4.4'F and a T0 of 66.0"F. Using emission factors in
Table 3-8 and the gasoline throughput associated with each model
plant allows the calculation of model plant emission estimates
for any geographical area. The equation for estimating model
plant emissions is as follows:
3-29
-------
TABLE 3.8. VOC EMISSIONS FROM REFUELING FOR
SERVICE STATION MODEL PLANTS8
Service station
Model Plantsb
Model Plant 1
Model Plant 2
Model Plant 3
Model Plant 4
Model Plant 5
Average
Throughput
Liters/Month
23,000
76,000
132,000
234,000
700,000
Average
Emission
Factor
mg/literc
1,340
1,340
1,340
1,340
1,340
Model
Plant
Emissions
Mg/yr
0.4
1.2
2.1
3.9
11.2
8 Not including emissions associated with spillage and tank
emptying/breathing.
b Model plants described in Chapter 2.
c Average emission factor based on the following:
RVP 11.4
Dispensed fuel temp. 66.0
Dispensed fuel/fuel 4.4
tank temp. diff.
3-30
-------
MPE = (Er) (MPT) (12 months/year)/(109mg/Mg)
where:
MPE = Model plant emissions, Mg VOC/yr
Er = Emission rate, mg VOC/liter
MPT = Model plant gasoline throughput, liters/month
3-31
-------
3,6. REFERENCES
1. Furey, Robert and Bernard Nagel. Composition of Vapor
Emitted from a Vehicle Gasoline Tank During Refueling.
Society of Automobile Engineers (SAE) Technical Papers
Series 860086. February 1986.
2. Gasoline Marketing Industry (Stage I) - Background
Information for Proposed standards, Preliminary Draft.
U.S. Environmental Protection Agency, Research Triangle
Park, NC. November 1991.
3. Evaluation of Air Pollution Regulatory Strategies for
Gasoline Marketing Industry. U.S. Environmental
Protection Agency, Office of Air Quality Planning and
Standards and Office of Mobile Sources. Ann Arbor, MI.
Publication No. EPA-450/4-84-012a. July 1984.
4. Draft Regulatory Impact Analysis: Proposed Refueling
Emission Regulations for Gasoline-Fueled Motor Vehicles
— Volume I - Analysis of Gasoline Marketing Regulatory
Strategies. U.S. Environmental Protection Agency.
Office of Air Quality Planning and Standards and Office
of Mobile Sources. Ann Arbor, MI. Publication No.
EPA-450/3-87-001a. July 1987.
5. Pacific Environmental Services, Inc. Description of
Analysis Conducted to Estimate Impacts of Benzene
Emissions from Stage I Gasoline Marketing Sources.
Prepared for U.S. Environmental Protection Agency.
Research Triangle Park, NC 27711. August 1989,
6. Design Criteria for Stage I Vapor Control Systems
Gasoline Service Stations. U.S. Environmental
Protection Agency, Office of Air Quality Planning and
Standards, Research Triangle Park, NC 27711. November
1975.
7. Norton, R.L., R.R. Sakaida, and M.M. Yamada (Pacific
Environmental Services, Inc.). Hydrocarbon Control
Strategies for Gasoline Marketing Operations. Prepared
for U.S. Environmental Protection Agency. State
Publication No. EPA-450/3-78-017. April 1978.
8. Telecon. Bowen, E., Pacific Environmental Services,
Inc. with Bradt, R., Hirt Engineers, U.S. Environmental
Protection Agency, September 25, 1991. Comments on
preliminary draft technical guidance document.
9. Letter, from Kunaniec, K., Bay Area Air Quality
Management District, to Shedd, S., U.S. Environmental
Protection Agency. July 31, 1991. Comments on
preliminary draft technical guidance document.
3-32
-------
10. Rothman, David, and Robert Johnson. Technical Report -
Refueling Emissions from Uncontrolled Vehicles.
Prepared for U.S. Environmental Protection Agency,
Office of Mobile Sources. Ann Arbor, MI. Publication
No. EPA-AA-ADSB-85-6, June 1985.
11. Motor Gasolines, Summer 1990 and Winter 1989-1990.
National Institute for Petroleum and Energy Research.
Bartlesville, Oklahoma. February 1991 and June 1990.
12. Volatility Regulations for Gasoline and Alcohol Blends
Sold in Calendar Years 1991 and Beyond. Federal
Register. Vol. 55, No. 112, 23658. June 11, 1990.
13. Clean fuels rules and guidelines negotiated rulemaking
advisory committee. Federal Register. Vol. 56, No.
103, 24157. May 29, 1991.
14 Fax communication to Norton, Robert, Pacific
Environmental Services, Inc., from Johnson, Robert,
U.S. Environmental Protection Agency, Office of Mobile
Sources. April 10, 1991.
15. Reference 12.
16. McAnnally, Michael and J.L. Dickerman (Radian
Corporation). Summary and Analysis of Data from
Gasoline Temperature Survey Conducted by American
Petroleum Institute. May 1976.
17. Reference 10.
18. Compilation of Air Pollutant Emission Factors, Fourth
Edition (AP-42). Section 4.4 Transportation and
Marketing of Petroleum Liquids. September 1985.
19. Reference 3.
20. Reference 4.
21. Procedures for the Preparation of Emission Inventories
for Carbon Monoxide and Precursors of Ozone Volume I:
General Guidance for Stationary Sources. Prepared for
U.S. Environmental Protection Agency, Research Triangle
Park, NC. Publication No. EPA-450/4-91-016. May 1991.
22. User's Guide to MOBILE4.1. EPA-AA-TEB-91-01. U.S.
Environmental Protection Agency. Office of Mobile
Sources. Ann Arbor, MI. July, 1991.
3-33
-------
23. A Survey and Analysis of Liquid Gasoline Released to
the Environment During Vehicle Refueling at Service
Stations. American Petroleum Institute. API
Publication No. 4498. June 1989.
24. Memorandum from Fricker, Robert L., California Air
Resources Board to Morgester, James T., California Air
Resources Board. April 18, 1991. Investigation of
Gasoline Spillage at Retail Service Stations.
25. Fax transmission from Kunaniec, K. Bay Area Air Quality
Management District, to Norton, R., Pacific
Environmental Services, Inc. September 19, 1991. Data
from Bay Area Spillage Study.
26. Stage II Survey Statistical Data Report. Lundberg
Reports Corporation. Fredericksburg, VA. August 1987.
27. Reference 8.
28. Reference 9.
29. Burlin, R., and A. Fudiruch. Air Pollution From
Filling Underground Gasoline Storage Tanks. Los
Angeles Air Pollution Control District. December 1962.
30. Nichols, R.A. Service Station Underground Tank
Breathing Emissions. R.A. Nichols Engineering.
October 13, 1987.
31. Memorandum. Simeroth, D.C., California Air Resources
Board, to Cackette, T., California Air Resources Board,
September 15, 1987. Determination of Mass Emissions
from Underground Storage Tanks.
32. National Institute for Petroleum and Energy Research.
Evaporative Losses from a Vented Underground Gasoline
Storage Tank (with addendum discussing August 24, 1987
meeting with CARB). Undated.
33. Reference 32.
3-34
-------
4.0 CONTROL TECHNOLOGY
This Chapter provides a basic technical discussion of
Stage II technology and equipment. Phase II vapor recovery
is also used to describe this technology. However, this
document uses the terminology, "Stage II". While the
fundamental concept of Stage II vapor recovery is simple,
the practical application becomes quite complex. There are
many components that have small but important roles in Stage
II systems. The initial sections of this chapter discuss
the types of Stage II systems and the system components.
Excessive equipment malfunctions and user
dissatisfaction have been traditional stumbling blocks to
Stage II program implementation. Where there were problems
with earlier generations of equipment a discussion of
corrections or improvements has been included.
Stage II originated in California and this State has
continued to be at the center of developing Stage II
technology. Fundamental to the Stage II program in
California (as well as the rest of the country) is the
equipment certification program conducted by the CARS. This
program is also discussed in the chapter. Much of the
information regarding system components and CARS
certification is taken from a paper presented at the 83rd
annual meeting of the Air and Waste Management Association
in June, 1990, entitled "Gasoline Vapor Recovery
Certification", by Laura McKinney of CARB.1
Finally, the chapter discusses the effectiveness of
Stage II systems. Results of studies of in-use
effectiveness and methodologies for determining program
effectiveness are provided.
4-1
-------
4.1 TYPES OF STAGE II SYSTEMS
Loading losses due to the refueling of motor vehicles
can be significantly reduced by Stage II systems. There are
currently two basic types of Stage II systems in use in the
United States. These are the vapor balance system and the
vacuum assist system.
4.1.1 Vapor Balance System
The balance type vapor recovery system operates on the
principle of positive displacement during gasoline transfer
operations. Balance systems use pressure created in the
vehicle fuel tank by the incoming liquid gasoline and the
slight negative pressure created in the storage tank by the
departing liquid to transfer the vapors through the
combination fuel dispensing/vapor collection nozzle, through
the vapor passage, and into the service station underground
storage tank. Because a slight pressure is generally
created at the nozzle/fillpipe interface, effective
operation requires that a tight seal be made at the
interface during vehicle fuelings to minimize vapor leakage
into the atmosphere. Also, it is very important that the
vapor path remain unobstructed.
The basic design of a balance system is shown in Figure
4-1. As illustrated, the vapors and liquid are simply
"balanced" between the vehicle and underground storage
tanks.
4.1.2 Vacuum Assist System
An assist system is designed to enhance vapor recovery
at the nozzle/fillpipe interface by drawing in vapors using
a vacuum. Because of this design, assist systems can
recover vapors effectively without a tight seal at the
nozzle/fillpipe interface. There are four assist systems
that are currently available and certified by the California
Air Resources Board (GARB): the Hasstech, the Healy, the
Hirt, and the Amoco Bellowless Nozzle Systems. The Hirt and
Hasstech Systems have a vacuum-generating device, such as a
4-2
-------
*>.
I
Figure 4-1. Vapor Balance System
-------
compressor or turbine that creates a vacuum such that vapors
are pulled from the vehicle tank into the storage tank.
They utilize a processing unit for combustion of the excess
vapor, while the Healy system creates a vacuum by spraying
liquid gasoline through saturated vapor by way of a jet, or
multi-jet pump, and the vapor is driven back to the
underground storage tank. A vacuum is created in the
bellowless system by a hydraulic pump driven by the
dispensed gasoline. The excess vapors are drawn through a
coaxial spout on the nozzle. The Red Jacket aspirator assist
,system was one of the first true aspirator assist systems to
be certified, but is no longer produced. It was fully
equipped with an aspirator, a modulating valve, and a check
valve; but it has not been sold since the early 1980's.
The Hasstech System, shown in Figure 4-2, uses a blower
as a vacuum generating device that is activated whenever
gasoline is dispensed. As product is dispensed, the vapors
are drawn through the vapor hose until they encounter a
valve that is located inside the dispenser. The purpose of
this valve is to prevent ambient air flow into the vapor
recovery line while other nozzles are in use. Vapors pass
through the valve, then through the blower located between
the dispensers and the storage tanks. This blower is
capable of a pressure differential of 20 inches water column
(in we), which means that the blower readily pushes the
vapor into the tanks. When there is an excess volume of
vapor from either Stage II or Stage I, the tank pressure
rises. When the pressure reaches approximately 1 in we, a
switch within the processor is activated and this initiates
processor operation. The processor incinerates the excess,
then automatically turns off when pressure equilibrium is
restored. This system is closed with a pressure/vacvium
relief valve on the tank vents. There is also a pressure
gauge located on the vent line that allows the
owner/operator to monitor the pressure of the system.
4-4
-------
PVVent
i
en
Figure 4-2. Hasstech Assist System
-------
The Hirt System is a vacuum assisted, vapor processing
Stage I-Stage II control system, shown in Figure 4-3, The
system is piped as a balance system, returning vapor from
nozzles to storage tank free space through unobstructed
vapor piping. Mi assisting vacuum is held in the storage
tanks by a vapor processor. The processor is piped into the
top of the storage tank vents which are manifold together
and closed from the atmosphere. The processor contains a
regenerative vapor turbine which prevents pressurizing by
removing excess vapor to the balancing forces, and a thermal
oxidizer which destroys only that vapor. If for any reason
the processor should shut down, the system will function as
a normal vapor balance system. The processor is
automatically activated if the vacuum degenerates to neat
atmospheric and remains activated until the vacuum reaches
about 0.5 in we.
Another example of an assist system is the Healy System
as shown in Figure 4-4. This system operates under negative
pressure derived from a gasoline driven jet pump.
Originally the jet pumps were located in the dispensers,
however, the newer system pumps may be in the vapor return
piping at the storage tank. The unit located at the tank is
called a multi-jet or mini-jet, depending on the number of
jet pumps it contains. The jet pump draws a strong vacuum
that creates enough suction to draw any excess liquid that
may be present in the vapor passage. When the pump switch
is activated, gasoline under pressure is provided to the jet
pump. At this point an internal pressure sensing valve
opens and a small stream of gasoline flows through the jet
pump back to the underground storage tank. Vacuum produced
by the mini-jet is immediately produced at a controlled
maximum level (15 to 70" we). When the nozzle is in use the
vapors are recovered through the jet pump and returned to
the gasoline storage tank. A vacuum regulatory, which has a
4-6
-------
PVVenl
Processor
with
Turbine
Figure 4-3. Hirt Assist System
-------
§
*J
a
>i
CO
*J
(0
•H
a
CO
id
0)
K
I
**
Q>
4-8
-------
0.17 inch diameter orifice, is located within the nozzle and
monitors this pressure. This opens into a pressure
regulated chamber that adjusts the flow of vapors and air
through the vapor recovery line. The regulator in the
nozzle is designed to open the vapor path when pressure
inside the nozzle is slightly above atmospheric pressure.
It also closes the vapor path when the pressure becomes
slightly negative; and this prevents excess air from
entering the system. This also keeps a slight vacuum (0 to
.25 in we) at the nozzle/fillpipe interface and a tight seal
is not necessary between the vehicle tank and the nozzle
fillpipe. Because of this pressure regulator, the high
vacuum in the vapor return line is not at the
nozzle/fillpipe interface. There is no need for an
incineration device with this system, because the amount of
ambient air drawn into the system is kept at a minimum.
Healy Systems which have the mini-jet or multi-jet unit are
required to have a pressure/vacuum (P/V) valve on the vent
pipes. The pressure setting of this valve is 1" we.
Another type of vacuum system, the bellowless nozzle
system shown in Figure 4-5, develops suction by a dual
chamber gasoline driven vacuum pump. Currently, the only
certification for a bellowless nozzle has been issued to
Amoco Oil Company. A vacuum is created by a hydraulic pump
driven by the dispensed gasoline. The vapors are drawn
through spout openings in a bellowless nozzle into the
underground tank. The vacuum is regulated by the flow of
fuel, and the ratio of gasoline dispensed to vapors
collected is approximately one-to-one. Because the vapors
are drawn into the tank at this "one-to-one" ratio, excess
vapors are not generated, and incineration is not necessary.
In addition, the vapor does not contact the liquid driving
the pump, thus not creating additional evaporation or
misting into the air-vapor mixture. The bellowless system
also has a P/V valve on the vent pipes. The current
4-9
-------
PVVENT
Figure 4-5. Amoco Bellowl ess Nozzle System
-------
settings are 8 oz. and -1/2 oz. The system is under
additional testing and may be certified by CARE with
different settings but a P/V valve will be required.
4.2 SYSTEM COMPONENTS
A more complete understanding of the technology of
Stage II vapor recovery may be gained by considering the
equipment on an individual component basis. In this
section, the function and operation of components are
discussed as well as a presentation of traditional problems
and improvements made to Stage II equipment.
**2.1 Vapor Recovery Nozzles
The collection of gasoline vapors at the vehicle-
fillpipe interface is the starting point for a Stage II
vapor recovery system. The component vital to this step is
the combination fuel dispensing/vapor collection nozzle.
The nozzle is responsible for dispensing gasoline into the
vehicle fuel tank while simultaneously collecting the vapors
being forced from the tank and routing them through the
vapor recovery hose and the underground piping to the
storage tank. Due to differences in Stage II vapor recovery
systems and the manner in which the vapors are collected,
the nozzles vary from vapor balance nozzles that require a
tight seal at the fillpipe interface to the "bellowless"
nozzle, which differs only slightly in appearance from
conventional nozzles. Figures 4-6 (balance), 4-7 (assist),
and 4-8 (bellowless) show various types of nozzles.
Many past problems with Stage II vapor recovery have
been associated with the vapor recovery nozzle. A survey
conducted in California in the late 1970's during the early
period of Stage II indicated that torn nozzle bellows and
faceplates, loose or missing latching lugs on balance
nozzles, loose or unwound latch springs on assist nozzles,
and fuel recirculation were among the most significant
problems.2 Also, a 1983 report to the California
legislature listed four major consumer complaints all of
4-11
-------
Secondary Shutotf
Vapor Check
Va I ve
Bellows
I
H
N)
Photos Courtesy of Dover Corporation/OPW Division Cincinnati, OH
and Emco Wheaton, Inc. Research Triangle Park, NC
Figure 4-6. Example Balance Nozzles
-------
BELLOWS SPRING
BELLOWS (BOOT)
^\XJ
\ \
LATCHING DEVICE-1 u SPOUT
Photo Courtesy of Emco Wneaton, Inc. Research Triangle Park, NC
-------
Figure 4-8. Example Bellowless Nozzle
* _1 A
-------
which were nozzle related: (1) spillage of liquid gasoline
during refueling, (2) equipment defects, (3) nozzle
operation and handling difficulty, and (4) gasoline
recirculation.3
Stage II equipment, especially nozzles, are far more
reliable and user friendly today than in the past. Hew
nozzles are shorter, narrower, and lighter than their
predecessors. Originally weighing over six pounds, newer
nozzle designs have reduced the weight by 2 to 3 pounds,4
rendering vapor recovery nozzles only slightly heavier than
conventional ones.
A major problem that occurred during the initial phase
of Stage II was the compatibility of Stage II nozzles and
vehicle fillpipes. There were many vehicles that had
fillpipes that simply would not accept the Stage II nozzles.
The State of California quickly recognized this problem and
passed legislation that required the standardization of all
vehicle fillpipes for 1977 and subsequent model years
(California Administrative Code, Title 17, Section 2290,
Chapter 7, page 267). Due to the difficulty of producing
cars with different fillpipes and to provide allow motorists
to fuel vehicles in all areas without difficulty, automakers
responded by standardizing vehicle fillpipes for vehicles
sold throughout the country.5 Therefore, newer model cars
should not have a problem using Stage II equipment, although
there will probably be a very small percentage of vehicles
still in use that have fillpipe configurations that will
make it difficult to use Stage II.
There are several parts of the nozzle which are
fundamental to the function of the nozzle and the recovery
of gasoline vapors. These parts are the bellows, the
primary and secondary shutoffs, the insertion interlock, the
latch assembly, the hold-open latch, and the vapor check
valve. Each of these units is discussed in detail in the
following.
4-15
-------
4.2.1.1 Nozzle Bellows. The nozzle bellows, or
"boot", is the device that captures the displaced gasoline
vapors at the vehicle fillpipe. Bellows were originally
composed of rubber-like materials over a shape-retaining
inner spring. The most recent generation of bellows are
made from shape-holding more durable materials (see Figures
4-6 and 4-7).
For balance systems, the tight fit at the vehicle
fillpipe interface is critical, so the bellows must be
compressed to create this seal. The faceplate and insertion
interlock (discussed later) are other components that assist
in assuring a tight fit. The tension of the bellows and the
difficulty of compression have been a source of consumer
complaints during the history of Stage II vapor recovery.
Also, the durability of bellows has been an often cited
problem.
The durability of bellows material has also been
significantly improved since the introduction of Stage II.
This is largely responsible for an increased life expectancy
of bellows for all systems and the improvements in the user-
friendliness of balance systems. The high spring-tension of
early balance bellows was responsible for much of the "hard-
to-use11 reputation of vapor recovery nozzles. The tension
the bellows exerts on balance-type faceplates is
substantially less now than it was years ago, and thc»
nozzles are consequently much easier to use.
The early popularity of assist systems was in part due
to the difference in the type of bellows necessary for
proper system operation. Because the vapors are drawn into
the bellows using a slight vacuum, a tight seal at the
vehicle fillpipe interface is not necessary. In fact, the
existence of a tight fit could cause removal problems and a
chance of pulling a vacuum on the vehicle tank. This less
stringent demand on the bellows allowed the use of lighter,
more pliable bellows material for assist systems.
Therefore, assist systems were attractive due to their
4-16
-------
increased user friendliness over the early designs of
balance nozzles. Improved technology has resulted in
lighter and more durable assist bellows, but the gap in user
friendliness has been closed by the improvements to balance
systems bellows.
Despite these improvements, the importance of the
bellows and the desire to avoid bellows maintenance continue
to interest the industry. This is evident in the excitement
and anticipation created by the bellowless nozzle (see
Figure 4-7). While the bellows improvements have lessened
many problems, the bellowless nozzle, in theory, will
eliminate the maintenance associated with nozzle bellows.
However, this bellowless nozzle has not been installed on a
wide scale basis at this point.
Part of the reason that this system is not currently
more prevalent is due to the fact that the system design was
developed by Amoco Oil company, and Amoco does not market
their gasoline products in California. Therefore, the
incentive to develop and fully market this product in the
past has not been great. However, due to the onset of Stage
II regulations in other areas marketed by Amoco, these
systems have been installed at approximately 100 stations in
St. Louis, D.C., and Philadelphia, with some "experimental"
sites in Maryland. There has been one bellowless nozzle
system certified by GARB for limited application with
certification testing for a second generation nozzle planned
for the near future. It is expected that with the
regulation in Dade County, the number of operating Amoco
bellowless nozzle systems could double due to the numerous
affiliated stations in this county. An Amoco representative
indicated that the initial plans are to limit the
availability of these systems to Amoco stations, although
there is the possibility that market rights will be sold to
other distributors in areas not marketed by Amoco (such as
California).6
4-17
-------
4.2.1.2 Faceplate or Facecone. Balance nozzles have a
tight-fitting soft donut-type faceplate, while assist
nozzles are often equipped with loose-fitting facecones.
The faceplates are designed to achieve the close seal
between the nozzle and vehicle fillpipe on which the balance
system depends. Assist facecones are loose-fitting and
often contain grooves to prevent a tight seal so that a
dangerous vacuum in the vehicle tank will be avoided. The
differences between balance faceplates and assist facecones
are apparent in Figures 4-6 and 4-7. There are exceptions
to this generic characterization. For example, one vacuum
assist system was originally designed and still can be used
with normal balance nozzles.
Difficulties have also been noted regarding the
durability of faceplates and facecones. New materials have
been developed which make these components stronger and much
more durable than their predecessors.
4-2.1.3 Primary and Secondary Liquid Shutoffs.
Conventional and vapor recovery nozzles have a primary
overfill shutoff mechanism, sometimes called the liquid
shutoff. This causes the nozzle to stop dispensing, thus
preventing overfills, when a sensing mechanism in the tip of
the spout (see Figure 4-6) detects that the spout tip is
submerged. A small tube inside the spout provides a path
for vapors from a small hole in the tip of the spout to a
chamber at the base of the spout. As gasoline flows through
the nozzle, vapor is sucked through this tube and fed
through tiny holes in the base of the spout back into the
gasoline. The suction that causes this is created by the
venturi effect of gasoline flowing through the spout. As
long as the flow of vapor is uninterrupted, the nozzle
continues to dispense gasoline. When the tip of the spout
becomes covered with liquid, however, the flow of vapors
stops and a vacuum is created. This vacuum pulls a thin,
rubber-like diaphragm and triggers a mechanical shutoff
mechanism to stop the flow of gasoline in the nozzle. The
4-18
-------
location of the diaphragm and the way it triggers the
shutoff differ with nozzle design.
Some nozzles have a three-ball latch mechanism that
causes the nozzle to shut off when the tip of the spout is
in liquid. Another type of shutoff mechanism uses the
vacuum to pull the diaphragm and two metal rollers away from
the shaft, which activates the shutoff.
If the primary shutoff fails on a conventional nozzle
the customer or attendant can easily recognize an overfill
situation as gasoline rises in the fillpipe or spills on the
ground. However, since vapor balance nozzles form a tight
fit at the fillpipe, it is difficult to determine if the
primary shutoff is malfunctioning. The nozzle may collect
the liquid, thus preventing a spill but allowing liquid to
collect in the vapor passage of the hose. Another common
problem for vapor balance and assist systems is "topping
off". Customers or attendants wish to fill the vehicle tank
as full as possible so they attempt to add more gasoline
once the primary shutoff has been activated. This provides
the opportunity for liquid to be introduced into the vapor
passage of the hose.
Because the balance system depends on a tight
nozzle/fillneck connection, there is a potential for
building up pressure in the vehicle tank if the vapor return
becomes blocked. This was a problem with the early designed
nozzles as pressure caused forcible ejection of liquid
product when the nozzle was removed at the end of the
fueling. To prevent this from occurring, a pressure sensing
shutoff mechanism (secondary shutoff) was required. The
pressure shutoff will be triggered if the primary shutoff
fails and the vapor line becomes blocked with fuel.
The secondary, or high-pressure, shutoff is required to
ensure that high pressure in the vehicle tank will not occur
when the vapor passage is blocked. The first vapor recovery
nozzles were required to shut off at about 20 inches water
column. This was later changed and nozzles are now required
4-19
-------
to shut off at or below 10 inches. The current industry
standard is 6 to 10 inches water for the pressure shutoff.7
Blockage of the vapor return path because of liquid, a
kinked or flattened hose or other obstruction, can cause the
nozzles to repeatedly shut off as pressure in the vehicle
tank builds up.
The secondary shutoff also acts as a guard against
recirculation of gasoline through the vapor passage. In the
event of a failure by the primary shutoff system, the build
up of liquid in the vapor passage will activate the
secondary shut-off and turn off the nozzle so that no
gasoline could be recirculated into the underground storage
tank. California Weights and Measures conducts stringent
testing of the secondary shutoff during nozzle
certification.
These secondary shutoffs have also contributed to the
hard-to-use reputation of balance nozzles. In most
instances continued shut offs occur when problems,
especially liquid blockage, exist in other parts of the
system, such as the vapor hose or the underground piping.
The certification process in California contains stringent
tests conducted by California Weights and Measures which
verify the delivery accuracy of nozzles and specifically
test the primary and secondary shutoffs (see Section
4.3.3.1).
4.2.1.4 Insertion Interlock Mechanism. As noted
previously, balance systems must maintain a tight fit at the
nozzle/fillpipe interface while assist and hybrid systems do
not. To achieve this tight fit, balance nozzles employ a
soft faceplate discussed above and an interlocking
mechanism. The insertion interlock, or "no seal-no flow"
device ensures that gasoline cannot be dispensed unless the
bellows of the balance nozzle are compressed to ensure a
tight fit at the nozzle/fillpipe interface. In some balance
nozzles, compression of the bellows opens a valve which
permits the flow of air from the spout-tip to the primary
4-20
-------
shutoff chamber. Attempting to dispense without compressing
the bellows therefore triggers the primary shutoff
mechanism. Other balance nozzles have a mechanical
interlock which prevents rollers from contacting the shaft
unless the bellows is compressed. The nozzle trigger is
loose and "floppy" until the bellows is compressed. This is
the type of interlock shown in Figure 4-6.
The difficulty in compressing the bellows so that the
insertion interlock will allow gasoline flow has been
another contributing factor to the complaints relating to
Stage II equipment. The earlier generation nozzles required
a pressure of up to twenty-four pounds to deactivate the
interlock. This, combined with the weight of the nozzle and
the tension of the springs in the bellows, made nozzle
operation difficult for many customers. However, the
improvement of each of these components has greatly reduced
this problem. The pressure required to deactivate insertion
interlocks has been decreased to as low as five pounds on
some nozzles.8
A lack of understanding of the interlock and latch
mechanisms can frustrate customers. This problem is one
that can be corrected with public awareness programs and
proper operating instructions at the pump.
4.2.1.5 Latch Assembly. The purpose of the latch
assembly is to allow the customer or operator to lock the
nozzle into the vehicle fillpipe by hooking the latch on the
lip of the fillpipe. The latch assembly may be a spring
wound around the spout, a ring around the spout (see Figure
4-7) or a bar riveted or screwed onto the spout (see Figure
4-6). This device is more critical to balance-type nozzles
because of the interlock and the greater tension exerted by
the bellows. Therefore, it is required on balance nozzles
and is optional for conventional nozzles and some assist
nozzles.
This simple device created problems specified in the
earlier surveys. The difficulties were mainly due to the
4-21
-------
latch assembly coming off the spout. Design and
manufacturing improvements have been made and complaints in
this area are now practically nonexistent.
4.2.1.6 Hold-Open Latch. This latch allows the nozzle
trigger to be "locked" in operating position, freeing the
operator to move away from the nozzle. Some establishments
elect to remove hold open latches for business reasons.
They prefer to keep customers at the nozzle so that they
will not leave vehicles unattended or drive off with the
nozzle still in the car. Hold open latches are not critical
to the actual recovery of vapors and nozzles are allowed
with and without them. The decision whether hold open
latches may be used is often decided by local fire marshals.
4.2.2 Vapor Check Valve
The vapor check valve opens and closes the vapor
passage between the underground tank and the atmosphere
(through the nozzle bellows). This valve closes when the
nozzle is not in use to prevent vapors from escaping. This
also prevents air leakage into the Stage II system and vapor
leakage out of it during vehicle refueling at another nozzle
or tank truck unloading. With the exception of a few
nozzles which have remotely-located flow-activated vapor
check valves, balance nozzles generally have vapor check
valves located in the nozzle at the base of the bellows
which are opened by compression of the bellows. Most assist
systems have vapor check valves located in the vapor passage
but not in the nozzle. For example, one assist system
nozzle has a ball-check valve (a very simple mechanism
involving a larger ball-bearing which blocks the vapor
passage when the nozzle spout points upward). Another has a
flow-control valve in the dispenser. Another system employs
a regulating diaphragm inside the nozzle designed to open or
close the vapor passage as necessary to minimize the
pressure difference inside and outside the nozzle.
4-22
-------
4-2.4 Hoses ,_a_nd_Hose Configuration Systems
4.2.4.1 poses. Vapor recovery hoses may be coaxial or
dual hose. Coaxial hoses contain two passages, configured
as a hose within a hose. One of the passages dispenses
liquid gasoline. The other passage, the vapor hose,
receives the gasoline vapors and carries them back through
the underground piping to the underground storage tank.
Most coaxial systems employ a 1/2 or 5/8 inch product hose
inside 1-^ to 1-% inch vapor hose. The single exception is
The Healy system which has the vapor hose inside the product
hose. Dual hose systems have separate hoses for the liquid
and vapor. Since 1986, all new or modified balance systems
installed in California must be coaxial. Other areas with
recently implemented Stage II programs only allow coaxial
hoses.
Historically, hoses have been a source of problems,
specifically with regard to their weight, durability, and
propensity to kink. Also, hoses often touched the ground
which made them susceptible to damage due to vehicles
running over them. Also, since Stage II was a technology
originally developed in central and southern California, the
durability of hoses in colder climates has been a concern.
The original two hose system was heavy and proved to be
awkward (due to hose twisting, etc.) for consumers and gas
pump attendants to use. To overcome this problem,
manufacturers developed a more manageable coaxial hose.
However, these were still hardwalled and continue to have a
weight problem. A second generation of coaxial hoses was
then developed that is much lighter and even more
manageable. The swivels that were necessary with the dual
hose systems and the hardwall coaxial hoses are not required
with these newer coaxial hoses. This further reduces the
weight of the hose and makes them easier to handle. Due to
improvements in thermal plastic materials, new coaxial hoses
will weight only about five pounds, which is comparable to
the weight of conventional gasoline dispensing hoses.
4-23
-------
Also, the durability of early model hoses under extreme
winter temperatures has been questioned. Fifth generation
coaxial hoses and bellows are designed to withstand
temperatures as low as -60° F.9 Stage II systems have been
installed in New Jersey and New York and no significant
increase in weather related defects has been observed.10'11
4.2.4.2 High Hana and Hose Retractor Systems. .Another
hardware improvement is the development of high-hang hose
dispensers and hose retractor systems. A major advantage of
these configurations is that they minimize hose kinks and
the possibility of the hose being flattened and help to
lessen the weight of the nozzle for the customer. This
helps to eliminate these situations which interfere with the
flow of gasoline vapors to the underground storage tank.
The hose retractor configurations also are designed to allow
any liquid gasoline trapped in the vapor portion of the hose
to drain into the fuel tank during normal fueling. The
exception to this are systems required to have liquid
removal devices. Figure 4-9 shows high hang hose and hose
retractor configurations.
4.2.4.3 Liquid Removal Systems. As stated above, one
major reason for the advent of the hose retractor systems
was to allow any gasoline trapped in the vapor passage of
the hose to drain into the fuel tank. However, the
structure of multiproduct dispensers does not contain the
loop that allows this drainage to the vehicle fuel tank.
Therefore, a method for removing liquid trapped in the vapor
passage of the hose was developed. Liquid removal systems
are designed to evacuate trapped liquid from the vapor
passages in coaxial hoses. They operate using the venturi
principle. A slight vacuum is created by the fuel flowing
in the interior hose that draws the liquid out of the vapor
passage and into the liquid gasoline stream. The venturi
device can be located at the dispenser end of the hose or
the nozzle end, depending on the type. Figure 4-10 shows an
example liquid removal device and illustrates its operation.
4-24
-------
Overhead Hose
Retractor
Breakaway
Coaxial
Venturi Located
Here Uses
Separate Liquid
Pick Up Below
Coaxial Hose
Hose Assemblies
Sloped To Permit
Natural Drainage
Into Vapor Return
Piping When
Retractor Is in
Retracted Posistior
Nozzle
Liquid Pick-Up Or
Venturi Placed
Here
Designed So That During
Fueling Hose Is Sloped
To Vehicle To Allow Any
Fuel In Vapor Line To
Drain Into Vehicle Fuel
Tank
Figure 4-9,
High Hang Hose Configurations
4-25
-------
4.2.4.4 Emergency Breakaways. An addition to Stage II
systems is the emergency breakaway valve. These breakaways
separate and close the product hose when a customer drives
off with the nozzle in the fillpipe, thereby preventing
damage to the equipment and reducing the danger of fire.
Figure 4-11 shows an example emergency breakaway.
4.2.5 Underground Vapor Piping
The underground vapor piping is an often ignored, but
important component of Stage II systems. In fact, GARB
certification includes not only the nozzles, hoses, and
other above ground equipment, but the underground piping as
well. Therefore, a CARB certified system must have the
correct underground piping configuration as specified in the
Executive Order.
The vapor piping begins with the riser pipe that is
located either inside the dispenser or on the pump island.
In many instances, this is a 3/4 inch galvanized riser pipe.
All vapor return and vent piping should be provided with
swing joints at the base of the riser to each dispenser, at
each tank connection, and at the base of the vent pipe riser
where it fastens to a building or other structure.
The underground vapor piping system can be made up of
individual return lines or a manifolded system. In either
instance, the minimum vapor pipe diameter is commonly 2 or 3
inches. The underground piping was originally all made of
steel, but fiberglass vapor piping has now become popular.
The individual return line system shown in Figure 4-12
is the simplest design and has one pipe for each underground
storage tank. If there are multiple dispensers of a
particular product or grade of gasoline, the vapor lines are
tied together into one line going to the appropriate tank.
Therefore, the vapors from the vehicle tank must be
transferred to the same tank from which liquid gasoline is
being drawn. The piping should slope towards the
underground storage tank with sufficient drop so that any
4-26
-------
Photo Courtesy of fioodyear T1re & Rubber, Co. Akron, OH,
MulthMoM Dispentw
ISUnd
p
fr
Coupling Assembly -
with Vtntuf i
Onpenaer End
- Coanial HOM AsWfntMy
| Product
D v»p°r
SD Product & Vapor
Splashback builds up
(vapor path blocked)
Intermittent nozzle
click-off
(product flow
interrupted)
Product
D Vapor
Product & Vapor
Product
flows without
interruption
Suction tube
Splashback removed
(vapor path cleared)
Photo Courtesy of Therrooid/HBD Industries, Inc. Bellefontaine, OH.
Figure 4-10 Example Liquid Removal Device
4-27
-------
Mporracowypeh Pmttm Mincing cMMMr
Photo Courtesy of HusKy Corporation Pacific, MO
Figure 4-11. Example Emergency Breakaway
4-28
-------
MOT 11
A1 IACM D If PIMM* STUt
UP PIPi *iOV» PUMP ISLAND.
Bit -JICTIOK TKitl PUMP IfLAWB
Figure 4-12.
individual vapor Balance System
underground Piping
4-29
-------
condensate or liquid in the vapor piping will drain to the
underground storage tank. Each tank also has a vent line
that is usually required to be at least 2 inches in
diameter. Therefore, there would be multiple vent lines
equal to the number of underground storage tanks. The vent
lines should also slope toward the tanks so that any
condensate will drain back into the tank.
In a manifold system, shown in Figure 4-13, all of the
vapor lines from the dispensers are linked to a common
manifold. This manifold can be run into a manifold box with
vapor connections to all of the tanks. More commonly, the
manifold is connected directly to the storage tank with
leaded gasoline, or the lowest grade of unleaded (in the
absence of leaded). This is to avoid contamination of the
higher grade gasolines. Again, the manifold must be sloped
adequately to allow any liquid present in the pipe to drain
to the liquid trap or storage tank. During vehicle fueling,
the vapors are returned to the appropriate tank due to the
slight vacuum created in the tank by the removal of the
liquid. As in the individual vapor return system, each
underground tank typically has a vent pipe.
The minimum height of the vent pipe off the ground is
usually determined by the Fire Marshal. A typical minimum
height is 12 feet above the adjacent ground level and should
vent upward or horizontally. Some areas allow pressure
vacuum vents on service station vent pipes. Pressure Vacuum
vents are required for some assist systems.
Problems can occur with the underground piping that
decrease the efficiency of the vapor recovery to very low
levels. These problems can take many forms from incorrect
piping size, to improper plumbing configurations where some
tanks are not even connected to the vapor piping system.
The most common problem associated with the underground
piping is the presence of low points in the line which allow
the build-up of liquid gasoline. Low points often occur due
4-30
-------
Figure 4-13
Manifolded Balance System Underground
Piping
4-31
-------
to inadequate backfilling of the piping or from running over
the piping by construction equipment prior to paving or
surfacing. Liquid blockage causes pressure build up which
either forces the vapors out at the nozzle/fillpipe
interface or causes the secondary shutoff mechanism to stop
the pumping of gasoline.
Many people with a great deal of experience with Stage
II systems believe that single most important element to a
Stage II program is to ensure that the systems are initially
installed correctly. Systems plumbed incorrectly reduce the
emission reduction potential of Stage II vapor recovery
substantially. Representatives in the San Diego Air
Pollution Control District of California estimate that the
underground piping at over 50 percent of the stations will
be installed improperly without an installation testing
program (these tests are discussed in Chapter 6 and
contained in Appendix I) and inspections to identify
improper systems.12
4.2.6 Abovectround storage tanks
With the problems associated with leaking underground
storage tanks and the resulting stringent UST and LUST
regulations, the interest in placing service station
gasoline storage tanks above ground is gaining attention.
In California there are a small number of service stations
that have Stage II systems on above ground storage tanks.13
For the most part, these are private card lock stations
serving fleets and small vaulted, tanks. Balance systems
have generally been installed for small tanks and vacuum
assist systems have been installed at these stations with
large bulk plant type tanks. The certification of above-
ground Stage II systems in California is discussed in
Section 4.3.5.
4-32
-------
4.3 CALIFORNIA CERTIFICATION PROGRAM
It is widely recognized and accepted that Stage II
technology originated in California and has developed
largely due to the regulations and requirements of the CARB
and local California air pollution agencies such as the Bay
Area Air Pollution Management District in San Francisco (Bay
Area), the South Coast Air Pollution Management District in
the Los Angeles area (South Coast), and the San Diego Air
Pollution Control District (San Diego). Many States and
local agencies in other parts of the country rely on
California for Stage II guidance due to their experience and
expertise.
California State law requires that the State Air
Resources Board adopt procedures for determining the compli-
ance of any system designed for the control of gasoline
vapor emissions during gasoline marketing operations.14 In
response to this legislative mandate, CARB developed proce-
dures and test methods which describe the requirements for
certification for all gasoline marketing emission sources.
Appendix C.I contains the requirements for certification.
Because it is not practical to test the efficiency of
the vapor recovery system in each service station, CARB
developed a "generic" equipment certification approach. In
this program a prototype Stage II vapor system is evaluated
and specifications developed. Systems that meet these
"certified" specifications may be installed without
individual efficiency tests.
CARB will accept applications for certification of
vapor recovery systems from any manufacturer, but there are
conditions which must be met by the manufacturer before
certification testing is initiated.15 The manufacturer is
required to demonstrate the ability to pay the costs of
testing prior to the commencement of CARB certification
testing. This demonstration may take the form of posting a
bond of not less than $20,000. In order to protect the
purchaser, CARB is also required to evaluate the adequacy of
4-33
-------
the planned methods of distribution and replacement parts
program, the financial responsibility of the applicant, and
other factors affecting the economic interests of the
eventual system purchaser. The manufacturer must also
provide a three-year warranty for the system. The only
exception to the warranty requirement is for those
components that the maintenance manual identifies as having
expected useful lives of less than three years, such as
vapor recovery nozzles. The warranty in these cases is
allowed to specify the expected life of the component.
Specifically, it is required that the application be in
writing, signed by an authorized representative of the
manufacturer, and include the following information:
1. A detailed description of the configuration of the
vapor recovery system which includes the
underground piping configuration and
specifications, the gasoline dispensing nozzle to
be used, engineering parameters for pumps and
vapor processing units, and allowable pressure
drops through the system.
2. Evidence demonstrating the vapor recovery
reliability of the system or device for 90 days.
The procedures by which this is determined are
discussed below in section 4.3.1.
3. A description of tests performed to determine
compliance with the general standards and the
results.
4. A statement of recommended maintenance procedures,
equipment performance checkout procedures, and
equipment necessary to assure that the vapor
recovery system, in operation, conforms to the
regulations, plus a description of the program for
training personnel for such maintenance, and the
proposed replacement parts program.
5. Six copies of the service and operating manuals
that will be supplied to the purchaser.
6. A statement that a vapor recovery system,
installed at an operating facility, will be
available for certification testing no later than
one month after submission of the application for
certification. The certification testing
procedure is discussed in detail in Section 4.3.2.
4-34
-------
7. The retail price of the system and an estimate of
the installation and yearly maintenance costs.
8. A copy of the warranty or warranties provided with
the system.
9. If the application is for a system previously
tested, but not certified, the application must
include identification of the system components
which have been changed, and any new test results
obtained by the applicant.
10. Any other information reasonably required by GARB.
While this list shows many requirements for certification,
the major portions of CARB requirements are the
operational/durability, or "90 day" test, and the
certification or "100 car" test.
4.3.1 Operational/Durability Test. "90 Day Test"
As stated above and contained in Appendix C.I, the
applicant must demonstrate the reliability of the system.
This demonstration is conducted by installing a system at an
operating station and observing the durability for at least
90 days.16 The facility utilized for certification testing
must have a minimum throughput of 100,000 gallons per month
and include at least six nozzles of each type submitted for
approval. No more than two types of nozzles can be present
at any one test facility. During this "operational" test,
replacement of components or alteration of the control
system is not allowed, except replacement or modification of
a component if it has been damaged due to an accident or
vandalism. No maintenance or adjustments to the system are
allowed during the test unless specifically called for in
the system's maintenance manual. The entire system is
sealed so that unauthorized maintenance or adjustment may be
detected. If detected, this can be reason for immediate
failure of the test. CARB observes the station frequently
during the testing period and evaluates the durability of
the system or components after this period.
4-35
-------
4.3.2 Certification Testing. "100 Car Test"
After meeting all other CARB requirements and
successful completion of the 90 day test, the efficiency of
the system is tested17 during at least 100 vehicle fuelings.
The test method is contained in Appendix C.2. The test
procedures provide for the fueling to occur during the
normal operation of the service station, but all CARB
efficiency testing is conducted in a self-service mode.
Before the 100 vehicle efficiency test can be conducted, the
entire vapor recovery system must be tested for leaks.
Each vehicle tank that is refueled is tested to
identify those which are leak tight. Vehicles that pass the
leak tight test may be included in the baseline population
if other measurements indicate that no vapors were lost
during the fueling operation.
Vehicle fuelings are observed until matrix requirements
are satisfied and at least forty baseline vehicles have been
identified. This matrix identifies vehicles by manufacturer
and year and ensures that the vehicles used during the test
are representative of the on-the-road vehicle population in
terms of vehicle miles travelled.
The test procedures for determining the efficiency of
systems to control gasoline vapors displaced during vehicle
fueling require that the weight of vapors collected at the
vehicle, corrected for vent losses, be compared to the
potential mass emission calculated for that vehicle. A
standard test sample of the vehicle population is tested and
an average efficiency calculated.
The potential mass emissions are determined during the
fueling of vehicles by measuring the mass of hydrocarbons
collected from vehicles from which no leak occurred
(baseline vehicles). Potential emissions are expressed as a
function of the vapor pressures of the dispensed fuel and
the temperature of the gasoline in the test vehicle tank.
The relationship is used as the baseline or reference from
4-36
-------
which the efficiency of a vehicle fueling vapor control
system is evaluated.
During these fuelings, spillage and spitback from the
system are also evaluated. Spillage is defined as "a loss of
more than one milliliter of liquid gasoline from the
gasoline nozzle as a result of preparing to fuel a vehicle
or at the end of a fueling operation in returning the nozzle
to the dispenser" and spitback defined as "a loss of more
than one milliliter of liquid gasoline during the dispensing
of gasoline.11 In order to pass this portion of the test, no
more than ten spitbacks or twenty instances of spillage per
100 vehicle fuelings can occur during the testing.
4.3.3 Approval of Other Agencies
The approval of three other State agencies is also
required as a precondition to GARB certification. State law
provides that the State Fire Marshal determine whether any
component of system creates a fire hazard.18 The Department
of Food and Agriculture, Division of Measurement Standards,
is given sole responsibility for the measurement accuracy
aspects, including gasoline recirculation, of any component
or system. Finally, the Division of Occupational Safety and
Health is designated the agency responsible for determining
whether any gasoline vapor control system or component
creates a safety hazard other than a fire hazard.19
Appendix C also contains regulations, requirements, and test
procedures for these other agencies.
4.3.3.1 California Measurement Standards Division.
Prior to Air Resources Board certification, the system must
be submitted for type approval to the California Department
of Food and Agriculture, Division of Measurement Standards
and certified by this division (see Appendix C.3).
The California Department of Food and Agriculture,
Division of Measurement Standards, issue certificates of
approval based on California Administrative Code Article 2,
Procedures for Type Approval Certification Evaluation and
Field Compliance Testing of Vapor Recovery Systems. This
4-37
-------
code establishes regulations to govern some design
characteristics of Stage II vapor recovery systems arid their
operation to ensure liquid recirculation is prevented.
There are several steps involved in order for
certification. It is the responsibility of the manufacturer
to request an application for the National Type Evaluation
Program (NTEP). Information regarding the design of the
system, including schematics, blueprints, instruction
manuals, brochures, and all other pertinent facts are sent
to the Director of the Measurement Division for a prelimi-
nary review.
Once the Director reviews the preliminary application
and approves, the applicant is authorized to install the
system in a prescribed location for use in the type approval
certification testing.
The Director, in conjunction with the County Sealer of
Weight and Measures for the designated location observe and
examine the system in operation normally within 30 to 90
days. During that time, one or more inspections will be
conducted which specifically relate the system components,
their performance, and their accuracy.
There are system installation specifications. There
must be a minimum of six nozzles installed on hoses of both
leaded and unleaded fuels, each tested a minimum of three
times during an examination. Prior to the field
examination, the dispenser meters for the test nozzles are
tested and adjusted accordingly.
Field compliance tests are conducted to examine: (1)
the proper operation of primary shut-off and secondary shut-
off devices, (2) the delivery accuracy of the system, and
(3) the performance accuracy of assist system evaporation
and volume change.
The test procedure for primary shut-off devices
involves filling the test unit with fuel dispensed from the
nozzle until the test unit becomes full. This should
activate the primary shut-off device. Ten consecutive
4-38
-------
override attempts are made which should result in automatic
nozzle shut-off before the dispenser volume indicator
increases more than 1/10 gallon limit. The 10 override
attempts are performed a minimum of three times for each
nozzle.
The secondary shut-off device is tested by introducing
sufficient fuel into the vapor return line to block the
return of vapors through the line. The nozzle and hose is
then held in a configuration so the liquid is concentrated
in the vapor section of the hose. Ten attempts are made to
dispense fuel into an empty test unit. The volume shown on
the dispenser indicator is recorded before and after each
attempt. The nozzle must shut off automatically before the
dispenser volume indicator increases more than 3/10 gallon
for each attempt. This procedure must be performed on a
minimum of 6 nozzles.
Compliance with delivery accuracy requirements is based
upon data recorded for at least 150 vehicles (formerly 300
vehicles) while observing customers fueling with the test
nozzles under normal field conditions. The 150 or more
vehicles should be representative of California vehicles.
The assist system evaporation and volume change
performance accuracy test is conducted because excessive
vacuum may result in artificial evaporation of customer
fuel. This would decrease the measured volume and also
cause possible implosion of vehicle fuel tanks.
In addition to all of these tests which are conducted
by Measurement Division personnel, type approval
certification is not issued until the applicant submits a
report of evaluation by an independent, pre-approved testing
laboratory. It is after review of all of the test data and
other information that the Division grants certification of
a vapor recovery system.
4.3.3.2 California Fire Marshal. Prior to Air
Resources Board certification of the vapor recovery system,
plans and specifications for the system must be submitted to
4-39
-------
the State Fire Marshal's Office for review to determine
whether the system creates a hazardous condition or is
contrary to adopted fire safety regulations (see Appendix
C.4). Final determination by the State Fire Marshal may be
contingent upon a review of each pilot installation of the
proposed system. The California Fire Marshall has
regulations, whose purpose is to establish minimum standards
of fire safety for vapor recovery systems or components.
Any manufacturer desiring certification and listing of
a gasoline vapor recovery system or component must submit a
completed application for evaluation and certification to
the State Fire Marshall. This form must be accompanied by
the proper fee. In addition, a test evaluation from a pre-
approved testing organization, as well as technical data and
black-line drawings suitable for reproduction must also be
submitted.
The final report should include failure analysis
engineering data, writing diagrams, operating and
maintenance manuals and photographs, together with a
description of the tests performed and the results. The
catalog number, the laboratory test report number, and date
should also be included.
After review and approval of the material, the Fire
Marshal issues a certification of the Stage II system. Each
vapor recovery system or component which is certified by the
California Fire Marshall must bear a label placed in a
conspicuous location and must be attached by the
manufacturer during production or fabrication.
4.3.3.3 California OSHA. Prior to certification of
the system, the manufacturer of the system must submit the
system to the California Occupational Safety and Health
Administration (Cal OSHA) for determining compliance with
appropriate safety regulations (see Appendix C.5). The
Division of Occupational Safety and Health of the Department
of Industrial Relations is the only agency responsible for
4-40
-------
determining whether a gasoline vapor control system or
component creates a safety hazard other than a fire hazard.
The General Industry Safety Orders (GISO) is the
guideline used in helping to make a determination. Each
section of the GISO relates to a different part of the
service station, ranging from the location of the storage
tanks to the safe operation of electrical equipment. All
electrical equipment and wiring must be installed in
accordance with the provisions of the California Electrical
Safety Orders. All electrical equipment integral with the
dispensing hose or nozzle must be suitable for use in the
proper locations.
They do not necessarily run tests, but assure that the
GISO guidelines and requirements and are met. The equipment
is tested by an outside lab which submits a report to
California OSHA.
The final determination is made when all of the
requirements have been met. A letter is sent to CARB
stating that the system in question has satisfied the
requirements of California OSHA.
4.3.4 Cost of Phase II certification
The certification of equipment is not an inexpensive
venture for equipment manufacturers. There are application
fees, government charges for testing, private laboratory
testing costs, as well as the manpower costs involved with
the oversight of the certification process. A fee not to
exceed the actual cost of certification is charged by the
Air Resources Board to each applicant who submits a system
for certification. A conservative estimate of the fees
charged by CARB is placed at around $5,000,20 excluding the
$20,000 bond discussed earlier. The contractor fee to
conduct the 100 car certification efficiency test has been
estimated at about $20,000.21 This puts the cost for only
the CARB portion of certification at approximately $25,000.
California State law allows the State Fire Marshal, the
Division of Measurement Standards, and the Division of
4-41
-------
Industrial Safety to charge reasonable fees for
certification of gasoline vapor systems not to exceed their
respective estimated costs. Payment of the fee is a
condition of certification. Representatives of major
equipment manufacturers estimate that the total cost for
obtaining GARB certification can range from $50,000 -
$100,000.22-23
4.3.5 Certification of Abovearound storage tank systems
Stage II systems have also been installed at gasoline
dispensing facilities with aboveground storage tanks. CARB
has certified several balance systems for small aboveground
vaulted tanks, as well as a Hirt assist system for similar
tanks. There are also Hirt and Hasstech assist systems
installed at bulk plant type card lock facilities, but no
certifications have been issued at this time. CARB
officials indicate that the certification of such systems on
a generic basis is expected in the future.24
Since most of these applications in California are at
private facilities, the conditions of the 100 car matrix
could never be met. Therefore, the certifications are based
on a combination of emissions monitoring, equipment testing,
and engineering analysis. Appendix D also contains examples
of executive orders for the small vaulted aboveground tanks.
4.3.6 Executive Orders
If the Executive Officer of CARB determines that a
vapor recovery system conforms to all requirements, an order
of certification, or Executive Order is issued. The Order
specifies the conditions which must be met by any system
installed under the certification. These specifications may
include the plumbing system, an equipment list, the vapor
hose configuration, and the maximum allowable pressure drop
through the system.
The interpretation of CARB executive orders can be both
confusing and frustrating. This is in part due to the fact
that many system updates and subsequent recertification of
the equipment occur. It is also due to the large number of
4-42
-------
components and manufacturers of these components. The
understanding of exactly what is "CARB certified" is not an
easy task, and areas with vapor recovery regulations which
rely on CARB certification should take the necessary time to
study and understand the Executive Orders. More discussion
on the determination of "approved systems" is given in
Chapter 6. Table 4-1 presents a list of current Stage II
CARB certifications and executive orders.
Appendix D contains a list of all Stage II CARB
executive orders issued since the initiation of the program.
This differs from Table 4-1 because some orders have been
updated, rescinded, etc.. Also included in the appendix are
summaries of the requirements for the most recent generation
of equipment. And finally, the appendix contains actual
executive orders. The executive orders provided include G-
70-52-AL that gives a summary of all above ground equipment
for Red Jacket, Hirt, and Balance systems? G-70-70-AB that
addresses the Healy aspirator assist system; G-70-7-AB that
addresses the Hasstech vacuum assist system; G-70-118 that
addresses the Amoco bellowless nozzle system; G-70-36-AC and
G-70-17-AB that have detailed descriptions of underground
piping requirements; and G-70-132 and G-70-133 that address
above ground tank systems.
If after certification of a system the manufacturer
wishes to modify the system, the proposed modifications must
be submitted again for approval. Such modifications may
include substitution of components, elimination of
components and modification of the system configuration and
may not require the full scale testing effort. If after
certification of a system, CARB finds the system to no
longer meet the specified certification specifications, they
may revoke or modify the prior certification.
4.3.7 Effectiveness of Systems
The test method for certifying Stage II systems states
that such a system "shall prevent emission to the atmosphere
of at least 90 percent or that percentage by weight of the
4-43
-------
TABLE 4-1. SUMMARY OF CARB EXECUTIVE ORDERS CERTIFYING
SYSTEMS TO BE AT LEAST 95 PERCENT EFFICIENT
Executive Order Title CARB Number
Certification of the Hasstech Model VCP-2 G-707-AB
and VCP-2A Phase II Vapor Recovery Systems
Relating to Modification of Certification of
the Emco Wheaton Balance Phase II Vapor
Recovery System G-70-17-AB
Recertification of the Exxon Balance Phase
II Vapor Recovery System G-70-23-AB
Recertification of the Atlantic Richfield
Balance Phase II Vapor Recovery System G-70-25-AA
Certification of the Modified Hirt VCS-200
Vacuum Assist Phase II Vapor Recovery System G-70-33-AB
Relating to Modification of Certification of
the OPW Balance Phase II Vapor Recovery G-70-36-AC
Recertification of the Texaco Balance Phase
II Vapor Recovery System G-70-38-AB
Recertification of the Mobile Oil Balance
Phase I Vapor Recovery System G-70-48-AA
Recertification of the Union Balance Phase
II Vapor Recovery System G-70-49-AA
Certification of components for Red Jacket,
Hirt, and Balance Phase II Vapor Recovery G-70-52-AM
Recertification of the Chevron Balance Phase
II Vapor Recovery System G-70-53-AA
Relating to the Certification of the Healy
Phase II Vapor Recovery System for Service
Stations G-70-70-AB
Certification of EZ-Flo Nozzle Company
Rebuilt Vapor Recovery Nozzles and Vapor
Recovery Nozzle Components G-70-78
Certification of EZ-Flo Nozzle Model 3006
and Model 3007 Vapor Recovery Nozzles and
Use of E-Z Flo Components with OPW Models
11VC and 11VE Vapor Recovery Nozzles G-70-101-B
4-44
-------
TABLE 4-1 (CONTINUED). SUMMARY OF GARB EXECUTIVE ORDERS
CERTIFYING SYSTEMS TO BE AT LEAST 95 PERCENT EFFICIENT
Executive Order Title CARB Number
Certification of Rainbow Petroleum Products
Model RA3003, RA3005, RA3006 and RA3007
Vapor Recovery Nozzles and Vapor Recovery
Components G-70-107
Certification of ConVault Incorporated
Aboveground Tank Filling/Dispensing Vapor
Recovery System G-70-116-A
Certification of Amoco V-l Vapor Recovery
System G-70-118
Certification of the Husky Model V Phase II
Balance Vapor Recovery Nozzles G-70-125
Certification of the OPW Model 111-V Phase
II Balance Vapor Recovery Nozzle G-70-127
Certification of the Bryant Fuel Systems
Aboveground Tank Filling/Dispensing Vapor G-70-128
Recovery System
Certification of the BRE Products, Inc.
Enviro-Vault Aboveground Tank
Filling/Dispensing Vapor Recovery System G-70-129
Certification of Sannipoli Corporation Petro
Vault Aboveground Tank Filling/Dispensing
Vapor Recovery System G-70-130
Certification of Hallmark Industries Tank
Vault Aboveground Tank Filling/Dispensing
Vapor Recovery System G-70-131
Certification of Trusco Tank, Inc.
Supervault Aboveground Storage Tank Filling/
Dispensing Vapor Recovery System G-70-132
Certification of LRS, Inc. Fuelmaster
Aboveground Storage Tank Filling/Dispensing
Vapor Recovery System G-70-133
Certification of the EZ-Flo Rebuilt A4000-
Series and HV-Series Vapor Recovery Nozzles G-70--134
Source: May 17, 1991 letter with attachments from James
Morgester, CARB, to Stephen Shedd, EPA.23
4-45
-------
gasoline vapors displaced during the filling of the
stationary storage tank as required by applicable air
pollution control district rules and regulations."26
Although this provides an efficiency of 90 percent, all of
the air pollution districts in California contain
regulations which require Stage II systems which achieve 95
percent efficiency.27 Therefore, CARB certifies systems as
95 percent efficient. In other words, a CARB certified
system has been tested and can be expected to achieved 95
percent or greater effectiveness in the removal of VOCs.
The systems shown in Table 4-1 have all been documented to
achieve 95 percent efficiency or better.
4.4 IN-USE EFFECTIVENESS
As stated previously, all Stage II systems certified in
California have been shown to operate with at least 95
percent removal efficiency. This efficiency is established
during the 100-car test segment of the certification
procedures. This 95 percent emission reduction is the
minimum required by districts in California and is required
by other States. However, after the equipment is installed
and normal operation begins, associated wear and tear,
malfunctions or system problems can result in reduction of
certified efficiency.
4.4,1 In-Use Efficiency
The term in-use efficiency is used to reflect the
actual average operating efficiency of the system. The in-
use efficiency takes into account system downtime,
malfunctions, and defects that can occur relating to
specific pieces of equipment. The in-use efficiency is
calculated by determining the frequency of specific
malfunctions and defects and assuming a specific efficiency
decrease associated with each malfunction or defect.
Factors affecting the in-use efficiency of a Stage II
system include:
4-46
-------
* misinstallation of aboveground or underground
equipment;
• specific nozzle defects or malfunctions;
* hoses tears, kinks, or liquid blockage;
« vacuum pump or vapor processor malfunctions; or
• generally poor maintenance.
Many defects or malfunctions to equipment are as a result of
the equipment being operated by the general public. As a
result, proper installation and maintenance of the equipment
is a crucial factor in keeping the in-use effectiveness as
close to 95 percent as possible.
Most of the discussion in this section describes the
affect on efficiency of defects in aboveground equipment.
Misinstallation of underground equipment can also cause
significant decreases in efficiency. One person interviewed
in California indicated that as much as 50 percent of the
facilities could have problems in underground piping
installations.26 This emphasizes the importance of
conducting the underground piping tests (liquid blockage,
backpressure, and pressure decay) to determine proper
installation. Chapter 6 discusses these tests in more
detail. Malfunctions or defective equipment left in
operation can significantly reduce the vapor capture and
hence the actual vapor reduction. Studies have shown that
the frequency of inspections made by enforcement personnel
can affect the in-use efficiency.29-30'31'32 More frequent
inspections will identify defective equipment, require
replacement of the equipment, and, as a result, improve
overall in-use efficiency.
4.4.2 In-Use Efficiency Calculations
Several pieces of data are necessary to calculate in-
use efficiency for a Stage II program. First is a database
of system malfunction and defects. This database is
necessary to establish the frequency of occurrence for
specific defects. Secondly, an efficiency decrease must be
4-47
-------
assigned to each malfunction or defect. This efficiency
decrease is an estimate of system efficiency decrease that
occurs with each malfunction or defect found. The overall
in-use efficiency is then the product of the individual
defect frequency and the efficiency decrease. The following
equation is used to calculate in-use efficiency.
E, - ET [lOO-^HED^HUOO-fFgHEDj)] --- [(100-
Fx) (EDX) ]
where :
E, - In-use efficiency, %
ET = Theoretical or certification efficiency, %
(typically 95 percent)
FX = Frequency of occurrence of defect x, %
EDX = Efficiency decrease assigned to defect x, %
Table 4-2 lists common defects for vapor balance systems and
the efficiency decrease associated with each defect. These
efficiency reductions have been developed and used by EPA in
previous in-use efficiency studies.53'34 The efficiency
decrease assumptions were in some cases obvious (i.e., no
vapor recovery installed resulted in 100 percent reduction
in efficiency) , while in other cases based on engineering
calculations (i.e., tears in nozzle boots). Appendix E of
this document contains an illustrative example of how to use
this data to generate an in-use efficiency estimate.
The example provided in Appendix E illustrates how
State or local agencies can use a database of defects to
estimate in-use efficiency. As new data becomes available,
efficiency decrease estimates in Table 4-2 can be refined to
better approximate efficiency reductions associated with
each defect, and a detailed database of malfunctions can be
obtained to estimate area specific in-use efficiencies. It
should be noted that the example calculations do not include
efficiency decreases due to underground piping problems.
For vacuum assist systems, malfunctions associated with
4-48
-------
TABLE 4-2.
EFFICIENCY DECREASES ASSOCIATED WITH STAGE II
BALANCE SYSTEM DEFECTS
Defect
Efficiency
Decrease
Assigned
(percent)
No Vapor Recovery Equipment Installed
(non-compliance)
- Facilities with no equipment on
any nozzle
- Facilities with at least some
vapor recovery
Nozzle Damage
Retractor Not Installed (all other V.R.
equipment installed)
Retractor Broken
Boot and Face Seal, or Boot Only, Not
Installed (V.R. nozzle installed)
Torn Boot
Face Seal Only Not Installed (remainder
of V.R. equipment installed)
Torn Seal
Vapor Hose Not Installed
Torn Vapor Hose
No Seal-No Flow Broken
Insufficient Hose Drainage
100
100
100
22
5
5
100
30
22
10
100
10
22
100
Source: 1987 RIA, Volume I, Appendix A.
4-49
-------
vacuum blowers and vapor processors would have to be
included.
4.4.3 Results and Conclusions
The in-use efficiency of a Stage II program is directly
proportional to proper installation, operation and
maintenance of the control equipment. Control agencies
where Stage II has been installed have asserted different
levels and frequencies of compliance inspections and
monitoring, and used public participation by complaint toll
free numbers to assure Stage II compliance. This section of
the document will focus on the end results of in-use
effectiveness estimates of Stage II systems and programs.
As discussed and described in the previous section,
surveys of installed equipment in areas with known levels of
compliance monitoring, and assumptions on the effect of
damaged or missing equipment, will allow the calculation of
the effectiveness of a Stage II program in a given area.
1PA has used this approach to calculate the effectiveness of
Stage II in previous studies for supporting an analysis of
Stage II versus onboard controls.35'36 These studies
calculated in-use efficiencies of 92 percent with semi-
annual inspections, 86 percent with annual inspections and
62 percent with minimal or less frequent inspections.
Figure 4-14 illustrates the relationship between inspection
frequency and in-use effectiveness. The range of inspection
frequencies shown on the graph is a simplification of actual
inspection frequencies and in most cases actual inspection
frequencies will fall between the data points.
EPA received a number of comments during the public
comment periods on the estimates shown in Figure 4-14.
Comments were received from auto manufacturers, control
agencies, equipment manufacturers, and oil company trade
associations that suggested both upward and downward
adjustments to the stage II in-use efficiency.37
The EPA evaluated new data in an effort to update the
in-use efficiency estimates and included this as Appendix A
4-50
-------
100
80
f
!
60
ui
H
20
0
Minimal
Annual
Frequency of Inspections
Semi-Annual
95
I
Certification
Figure 4-14
Relationship of Inspection Frequency to Program
In-Use Efficiency
-------
to the 1987 Draft RIA. As discussed previously in this
chapter, EPA also examined a recent report on inspection of
all Stage II service station installations in the
Washington, D.C. area, and revisions were subsequently made
to the estimates for the frequency and types of defects
affecting Stage II systems. Using this information, the
Agency's estimate for the lower end of the Stage II
efficiency range was adjusted from 56 to 62 percent.
The EPA also evaluated California Air Resources Board
data, which were presented in the 1983 Report to the
Legislature.38 An attempt was made to cull inspection data
dealing with only the newest Stage II systems. However, the
data were insufficient to differentiate between system type,
so no refinement of their 80-92 in-use efficiency rate could
be obtained. The analysis used the average of this range.
Additional data were obtained from randomly selected service
stations in the Bay Area of California, which indicated an
in-use efficiency of 90 to 92 percent; however, the data
were considered inadequate to update the in-use figure for
the entire State of California. Therefore, the upper end of
the in-use efficiency range used in the 1987 RIA was
maintained at 86 percent.
Since publication of 1987 RIA, additional data were
obtained that included inspection results about 12,000
nozzles in California.37 These inspections took place in
1986 and 1987 in San Diego, San Francisco Bay Area, and in
the South Coast (Los Angeles) areas of California. Based on
discussions with personnel in each of these areas, semi-
annual inspections would best represent their inspection
program (See Chapter 6). The data available allowed
comparison between older and newer nozzle equipment. The
results of these inspections indicated an overall in-use
efficiency of 92.5 percent for all nozzles, 92 percent for
older nozzles, and 94 percent for newer nozzle equipment.
The data from these inspections is used in Appendix E for
the illustrative example.
4-52
-------
Not taken in account in any of these in-use efficiency
calculations is misinstallation of underground vapor piping.
Figure 4-14 assumes 100 percent proper installation,
operation, and maintenance of belowground vapor piping
system.
In addition, Figure 4-14 presents only in-use
efficiency of controls if they are installed at 100 percent
of the dispensing facilities. Many areas may use size
exemptions. Table 4-3 summarizes the gasoline consumption
that would be exempted under different throughput level
cutoffs. These gasoline consumption levels were calculated
based on the size distribution information presented in
Chapter 2. Figure 4-15 presents in-use efficiency for the
different levels of exemptions. The curves are compared to
the information in Figure 4-14, that represented essentially
no exemptions.
In conclusion Figure 4-15 presents the range of in-use
effectiveness of Stage II programs and its relationship to
frequency of inspection and exemption levels. While it is
well documented that Stage II systems can achieve 95 percent
or better control efficiency, in-use efficiency is
demonstrated to drop significantly without proper
installation, operation, and maintenance by the owners and
operators.
4-53
-------
100
80
60
*.
I
Ul
40
20
0
None
Minimal
Noexemp
< 10.000, < SMDO
Program In-Use Efficiency
^"*~^^^ Frequency of
^""""•"••••^^Jiispectkws
Exemption Level ^^"~"~-~«^
No Exempt. —
2,000
10,000
10,000450,000 --
Minimal
62
61
60
56
Annual
86
84
84
77
Se mi-
Annual
92
90
89
83
Certifi-
cation
95
93
92
86
Annual
Frequency of Inspections
Semi-Annual
Certification
Figure 4-15.
Relationship of Inspection Frequency to Program
In-Use Efficiency with Exemptions
-------
TABLE 4-3. PERCENT CONSUMPTION EXCLUDED WITH VARIOUS
STAGE II EXEMPTION SCENARIOS
EXEMPTION SCENARIO
PERCENT CONSUMPTION
EXCLUDED PROM
REGULATION
EXEMPT STATIONS 2,000 GAL/MON
2.4%
EXEMPT STATIONS < 10,000 GAL/MON
2.8%
EXEMPT STATIONS < 10,000 GAL/MON AND
INDEPENDENTS < 50,000 GAL/MON
10.0%
Exemption values based on metropolitan area throughput by
model plant shown in Table 2-9, since most, if not all,
nonattainment areas are metropolitan areas. Table 2-10 was
used to estimate exemptions for independents. The following
assumptions were used:
< 2,000 gal/aon = Model Plant la
< 10,000 gal/mon = Model Plant 1
< 10,000 gal/mon non-independents, < 50,000 gal/mon
independents = Model Plant 1 plus independents in
Model Plants 2 and 3
4-55
-------
4.5 REFERENCES
1. MeKinney, L. California Air Resources Board.
Gasoline Vapor Recovery Certification. (Presented
at the Air and Waste Management Association 83rd
Annual Meeting. Pittsburgh, PA. June 24-29,
1990).
2. South Coast Air Quality Management District,
"Phase II Vapor Recovery Evaluation Program",
1979.
3. California Air Resources Board. A Report to the
Legislature on Gasoline Vapor Recovery systems for
Vehicle Refueling at Service Stations. 1983.
4. Massachusetts Division of Air Quality Control.
Stage II Gasoline Vapor Recovery Program
Background Information and Technical Support
Document. January 1989.
5. Telecon. Bowen, E., Pacific Environmental
Services, Inc., with Walker, G., Motor Vehicles
Manufacturers Association, and Brooks, D.,
Chrysler Corporation. October 31, 1991. Fillpipe
standardization.
6. Telecon. Norwood, P., Pacific Environmental
Services, Inc., to Strock, D., Amoco Research.
April 30, 1991. Amoco bellowless nozEle.
7. Reference 1.
8. Reference 1.
9. Reference 1.
10. Memorandum from Norwood, P., Pacific Environmental
Services, Inc. to Shedd, S., U.S. Environmental
Protection Agency, Chemicals and Petroleum Branch.
February 22, 1991. Trip Report to New Jersey
Department of Environmental Protection.
11. Memorandum from Norton, R., Pacific Environmental
Services, Inc. to Shedd, S., U.S. Environmental
Protection Agency, April 29, 1991. Trip Report to
New York Department of Environmental Conservation.
12. Memorandum from Norwood, P., Pacific Environmental
Services, Inc. to Shedd, S., U.S. Environmental
Protection Agency, April 30, 1991. Trip Report to
California Agencies to Discuss Stage II.
13. Reference 12.
4-56
-------
14. California Administrative Code, Section 41954.
15. State of California Air Resources Board,
Certification Procedures for Gasoline Vapor
Recovery at Service Stations, Method 2-2.
16. Reference 15.
17, State of California Air Resources Board, Test
procedures for determining the efficiency of
Gasoline Vapor Recovery systems at Service
Stations, Method 2-1.
18. California Administrative Code, Section 41955.
19. California Administrative Code, Section 41957.
20. California Air Resources Board Compliance
Division, Gasoline Vapor Recovery Certification
Program, Schedule of Test Fees. March 1991.
21. Reference 12.
22. Telecon. Norwood, P., Pacific Environmental
Services, Inc. to Parrish, D., Emco Wheaton.
February 15, 1991. GARB Certification program.
23. Telecon. Norwood, P., Pacific Environmental
Services, Inc. to Brown, B., OPW. April 13, 1991,
GARB Certification Procedures.
24. Telecon. Norwood, P., Pacific Environmental
Services, Inc. to Zimmerman, G., California Air
Resources Board. October 17, 1991. Aboveground
tank certifications.
25. Letter with attachments from Morgester, J.,
California Air Resources Board, to Shedd, S., U.S.
Environmental Protection Agency. May 17, 1991.
Information related to GARB Stage II program.
26. Reference 15.
27. Telecon. Norwood, P., Pacific Environmental
Services, Inc. to McKinney, L., California Air
Resources Board. May 13, 1991. Requirements of
GARB Efficiency Test.
28. Reference 12.
4-57
-------
29. Evaluation of Air Pollution Regulatory Strategies
for Gasoline Marketing Industry. U.S. Environ-
mental Protection Agency, Office of Air Quality
Planning and Standards and Office of Mobile
Sources. EPA-450/4~84-012a. July 1984.
30. Draft Regulatory Impact Analysis: Proposed
Refueling Emission Regulations for Gasoline-Fueled
Motor Vehicles — Volume I - Analysis of Gasoline
Marketing Regulatory Strategies. U.S. Environ-
mental Protection Agency. Office of Air Quality
Planning and Standards and Office of Mobile
Sources. EPA-450/3-87-001a. July 1987.
31. Reference 3.
32. sierra Research, An Analysis of Stage II and
Onboard Refueling Emissions Control, prepared for
Motor Vehicle Manufacturers Association, Inc.
November 30, 1988.
33. Reference 29.
34. Reference 30.
35. Reference 29.
36. Reference 30.
37. Evaluation of Air Pollution Regulatory strategies
for Gasoline Marketing Industry - Response to
Public Comments. U.S. Environmental Protection
Agency. Office of Air Quality Planning and
Standards and Office of Mobile Sources. EPA-
450/3-87-012C. July 1987.
38. Reference 3.
39. Inspection Summaries of California Air Resources
Board Phase II Vapor Recovery Inspections. August
1986 through October 1987. Received from
McKinney, L., CARS. October 1991.
4-58
-------
5.0 STAGE II COSTS
The purpose of this chapter is to present the costs
associated with the purchase, installation, and operation of
Stage II equipment. This cost information is useful to
State and local regulatory authorities when evaluating the
cost impacts or burdens of a proposed Stage II vapor
recovery program, and to weigh these cost impacts against
the emission reduction benefits achieved. In addition, this
information is useful when reviewing cost burdens presented
by commenters when implementing a Stage II program.
Developing and evaluating cost estimates for Stage II
systems has been a difficult task. EPA and industry have
evaluated unit costs using unit cost estimate approaches as
well as total cost estimate approaches from quotes from
stations that have recently installed and purchased Stage II
systems. In addition, these studies came at a time when
each study was trying to influence a decision between Stage
II and onboard refueling controls. These cost methods were
used and issued in a number of recent Stage II cost studies
by industry and EPA.
The unit cost estimate approach was based on model
station sizes and equipment specifications for all
components in a Stage II system. The cost of each piece of
necessary equipment was obtained, along with its
installation and maintenance costs. These costs were then
summed to produce a "ground-up" estimate of Stage II costs.
The total cost estimate approach, using cost quotes
surveyed from stations that have installed Stage II
equipment, is a simpler approach to obtaining Stage II
costs, but has many drawbacks. Some stations keep detailed
5-1
-------
cost records on Stage II installation while others will have
only the total cost. This makes comparison of costs very
difficult. Compounding this problem is most stations re-
model or replace storage tanks or dispensers at the same
time they are installing Stage II systems. These non-Stage
II costs can, in many cases, be much higher than Stage II
installations costs. Trying to compare a mixture of
detailed and non-detailed cost quotes, and attempting to
subtract out non-Stage II costs, can not only be difficult
and some times impossible to perform, but can add multiple
assumptions and uncertainties into what were once "actual"
Stage II costs. Without detailed costs it is also
impossible to analyze the reasons associated with any
outlier costs obtained from a total cost survey. This
chapter discusses and presents results of both cost
approaches, and compares all of the recent cost studies
performed or provided to EPA to provide the user with a
range of costs to use in their own assessment.
The costs presented in this chapter are divided into
aboveground and below-ground components. Aboveground
equipment consists of all the nozzles, hoses, swivels, check
valves, and other related components needed at the
dispensers to capture the vapors displaced during refueling.
The costs presented are limited only to equipment that has
been certified by the California Air Resources Board (CARB)
and is currently being marketed for Stage II systems. The
below-ground equipment consists of the piping needed to
route the vapors back to the underground tank. The
aboveground costs at a facility are driven by the number of
nozzles present at the service station, while underground
costs are driven by the physical layout of the facility.
Many times commenters will present Stage II costs on a
dollar per nozzle basis. But because underground costs are
not dependent on the number of nozzles, and because
underground costs can represent more than half of the Stage
II costs, reporting costs on a dollar per nozzle basis is
5-2
-------
not very useful. This report presents costs for the entire
vapor recovery system, broken down into aboveground and
below-ground components. Because there can be an infinite
number of service station configurations, costs are only
presented for model facilities (discussed in Chapter 2},
chosen to represent a cross section of the service station
industry.
Cost for key components (those having the biggest cost
impact and those requiring the most replacements) will be
discussed. Because most areas implementing Stage XI have
been taking advantage of the California certification
efforts by allowing only systems certified in California,
component costs are presented only for certified components.
In this chapter discussions of current equipment costs
for above and below ground components are presented. Also
presented is a discussion of capital and annual costs for
model facilities, a comparison of model plant costs with
several cost surveys conducted in St. Louis, and a
presentation of the latest 1991 Stage II cost estimates.
5.1 EQUIPMENT, INSTALLATION, AND ANNUAL COSTS
As discussed above, the costs are presented separately
for aboveground components and underground components. Also
presented in this section is a discussion of the impacts the
underground storage tank (UST) program could have on Stage
II implementation costs.
5.1.1 Aboveground Costs
The aboveground costs are associated with the hardware
necessary to capture the vapors displaced at the vehicle
fillneck during vehicle refueling. The discussions of unit
costs will be limited to certified components. Appendix D
contains a list of CARB's certified systems and a list of
the equipment specific to those systems. Most maintenance
items and replacement components are associated with the
aboveground equipment. The discussion in this section will
be more detailed for the higher cost, more maintenance
5-3
-------
intensive equipment (i.e., nozzles and hoses), and less
detailed for the lower cost, less maintenance intensive
equipment (i.e., swivels, check valves, etc.).
Costs presented in this chapter do not include costs
for the Amoco bellowless nozzle system. As discussed in
Chapter 4, full scale production of this system has not
occurred. An Amoco representative stated that the actual
installed costs once a wide spread production began could
not be estimated at this point.
5.1.1.1 Nozzles. The vapor recovery nozzles discussed
in Chapter 4 are the key to the vapor recovery capture.
Without a proper functioning and well maintained nozzle,
emissions capture can be almost zero. Appendix D lists the
nozzles approved for use for the balance, hybrid, and vacuum
assist systems. Information is presented for all configura-
tions and generations of nozzles, however the costs in this
section will be presented only for the latest equipment on
the market today. California maintains certification lists
for older generation equipment since many of these systems
are still being used. New Stage II programs, however, are
excluding most older equipment and are limiting acceptable
Stage II systems to those of the latest design. For
example, New York will allow only fourth generation or newer
vapor recovery components1 and St. Louis will allow only
coaxial nozzles and hoses and will not allow twin hose
configurations.2
The newest of the certified balance nozzles are the
A4005 from EMCO Wheaton, the 11IV from OPW and the Model V
from Husky. These are the only certified balance system
nozzles being offered by the original equipment
manufacturers. The cost for these nozzles and for vacuum
assist nozzles are comparable at about $240.3'4*5
Individually these cost seem small, but the costs can mount
up quickly when there a large number of nozzles, especially
if the station uses multi-product dispensers (the multi-
product dispenser for this document refers to a dispenser
5-4
-------
providing three products on each side of the dispenser, one
nozzle per product, resulting in six nozzles per dispenser).
The portions of the nozzle most susceptible to wear are
the nozzle faceplate and bellows. These are also key items
in the vapor capture system. These components cost about
$15 for the faceplate,6 and about $30-50 for a bellows
replacement kit.7'8 The life of the equipment will vary, but
a service station can expect, on average, to replace bellows
and faceplates about three times per year for balance
systems and two times per year for vacuum assisted systems.9
Other components in the nozzle (i.e., shutoff
mechanisms, no seal/no flow check valves, etc.) are more
difficult to repair. If these components fail, the nozzle
usually has to be replaced. The station operator can
replace the nozzles with new equipment at the cost stated
above or can reduce his costs by purchasing "rebuilt"
nozzles. Rebuilt nozzles use the same core but with new
components built inside. Nozzle manufacturers will rebuild
nozzles and sell them back at a reduced price. The
manufacturers buy back the cores of the used nozzles, repair
and resell them as certified nozzles. Core credits given by
the manufacturers are typically around $50. Rebuilt nozzle
costs range from $14510 to about $190.11
The State of New York only allows rebuilt nozzles
repaired by the original equipment manufacturer.
California, on the other hand, has certified rebuilt nozzles
by two rebuilding companies, Rainbow and EZ-flo, These
nozzles have been certified for use in balance system
installations. The cost of these nozzles are about $190.12
Table 5-1 summarizes the costs associated with purchase and
maintenance of Stage II vapor recovery nozzles.
5.1.1.2 Hoses. The original Stage II systems
incorporated a twin hose approach to vapor recovery. One
hose transferred the liquid, as in conventional vehicle
refueling, and an identical hose was used as a vapor return
hose. These hoses were relatively inexpensive at about
5-5
-------
TABLE 5-1. PURCHASE COSTS FOR VAPOR RECOVERY NOZZLES
AND REPLACEMENT PARTS3'4'5
(Kay 1991 Dollars)
Item Cost
Nozzle Costs
New Nozzle $240
Core Return Credit $50
Rebuilt Nozzle $190
Component Costs
Nozzle Boot $25
Boot Kit $40
Face Seal Kit $15
Clamp Kit $5
Boot Assembly Kit $30-50
5-6
-------
$30.13 However, the twin hose systems were very hard to
operate. Coaxial vapor return hoses eliminated the
difficulties caused by twin hoses but cost considerably
more. Coaxial systems represent the latest technology in
use in California and are required on all new installations.
They also are the only systems allowed in St.Louis, New
York, New Jersey and Dade County, FL. A wide variety of
materials and manufacturers are being offered for new Stage
XI coaxial hose systems. Manufacturers of certified coaxial
vapor recovery hoses include Goodrich, Goodyear, Dayco,
Gates and Thermoid. Hew hose materials make the latest
hoses more durable and, at the same time, more lightweight
and flexible. The costs for the coaxial hose range from
$140 to $230. u'15-16 (See Table 5-2.)
Hose life has been extended greatly because of the new
material, and because of the requirement for high hang hose
retractors or high hang dispensers. These requirements
force the hoses up off the ground and minimize or eliminate
hose problems such as collapsed hoses from being run over by
a vehicle, or hose tears and wearing from being constantly
scuffed on the ground. With the use of high hang hose
retractors or high hang dispensers, it is conservatively
assumed that vapor hose replacement would occur only on an
annual basis.
High hang hose retractors and high hang dispensers also
minimize vapor path blockage in the vapor hose caused by
spitback or by liquid condensation. For high-hang
multiproduct dispensers, venturi trap are required. These
liquid removal systems consist of a small tube inserted in
the vapor line extending to the low point of the hose. A
venturi is placed in the liquid delivery hose and dispensed
liquid passing through the venturi creates a vacuum in the
tube. This vacuum draws the liquid from the low point in
the hose into the liquid delivery hose. Liquid removal
systems can be purchased separately or in conjunction with a
coaxial hose assembly. These systems cost $200 if purchased
5-7
-------
separately17 or $240-$430 if purchased with a coaxial hose
assembly.18 Table 5-2 summarizes the costs associated with
vapor recovery hose purchase and replacement.
5.1.1.3 Other Components. Other components that must
be purchased with the aboveground equipment could include
high-retractor hose assemblies, hose breakaway fittings,
vapor check valves, swivels (nozzle, island, dispenser,
retractor), flow limiters, and hose splitters. Table 5-3
illustrates typical costs associated with these components.
These pieces of equipment are not expected to wear or fail
at the same rate as nozzles, bellows, faceplates, or hoses,
and are expected to operate relatively maintenance free.
5.1.1.4 pispenser Modi f i cations. Product dispensers
at existing service stations will have to be converted to
allow the installation of vapor return piping. Conventional
dispensers will typically have room within the dispenser to
allow the vapor piping riser to extend into the dispenser
and exit out the side. Newer dispensers, such as multi-
product dispensers, may have to be converted to allow the
installation of the vapor piping through the dispenser
housing and back into the underground piping. California
has included such dispenser modifications as part of a
certified system since the manner in which the piping is
plumbed through the dispenser could affect the backpressure
experienced in the vapor line at the nozzle, thereby
affecting the system's ability to recover the vapors.
Typical costs to modify an existing dispenser is about
$50-60.19
5.1.1.5 Vapor Processors. The Hirt and Hasstech GARB
certified vacuum assist systems use a thermal oxidizer as
the vapor destruction device. The thermal oxidizer system
necessary to transport vapors from the underground tank to
the vapor processor consist of a pilot/ignition system,
vapor pump, PV vents, etc. The cost of a vapor processing
system is about $4,000.20
5-8
-------
TABLE 5-2. TYPICAL VAPOR RECOVERY HOSE COSTS13'14'15
(May 1991 Dollars)
Item6 Costs
Coaxial Hose $140-$230
Liquid Removal System $200
Coaxial Hose with Removal System $240
8 Costs presented for a typical 10 foot hose system.
TABLE 5-3. TYPICAL COSTS OF OTHER VAPOR
RECOVERY COMPONENTS10'12'13
(May 1991 Dollars)
Item Costs
High hang hose assembly $100
Hose break away fittings $140
Vapor check valves $80
Swivels
Nozzle $60
Island $60
Dispenser $60
Retractor $60
Flow limiters
Hose splitters $60
5-9
-------
The vapor pump and the vapor processor will require
additional adjustments and repairs. It has been estimated
that annual maintenance costs would be as much as $400-600
per year.21
5.1,1.6 Installation. Installation of the aboveground
equipment consists of assembling the hoses, nozzles,
swivels, check valves, etc., and attaching the nozzle/hose
assembly to the vapor piping exiting the dispenser. It has
been estimated that installation would cost about $80 per
nozzle. If a vacuum assist unit is being installed an
additional $1,300 would be necessary to take care of the
thermal oxidizer and vapor pump installation.22 The Healy
System requires the installation of the jet pump used to
create the vacuum in the vapor return line. It has been
estimated that the installation of the jet pump would cost
$535.23
5.1.2 Underground Piping
The underground portion of the vapor recovery system
consists of all the underground piping and fittings
necessary to allow the captured vapors to be returned to the
underground storage tank. Costs of the underground
components are directly affected by the service station
configuration (i.e., the number of islands, the distance
between islands, the distance from the islands to the
underground tank), the type of system (individual balance
system, manifolded balance system, hybrid system, or vacuum
assist system) and other station physical characteristics
(amount of concrete over underground tanks, amount of
backfill material required, or whether the storage tanks are
located above the islands). The following subsections
discuss some of these costs in more detail.
5.1.2.1 Vapor Piping. Most vapor recovery piping
being used in recent installations consists of fiberglass
pipe. Reasons usually cited for using this type of piping
is that it is leak resistant, easy to work with, and easy to
install (i.e., glued not threaded). Typical vapor piping
5-10
-------
consists of 2 inch or 3 inch pipe laid in a trench, sloping
down to the underground tank. The amount of piping required
is certainly affected by specific facility distances, but
also whether individual or manifolded vapor piping is
used.24 Table 5-4 summarizes the piping differences between
a manifolded vapor balance system and an individual vapor
balance system. Vacuum assist systems can either be
manifolded or individual. Table 5-5 summarizes the piping
costs for different certified systems assumed for a typical
9 nozzle, 65,000 gallon per month service station.
5.1.2.2 Trenching and Backfilling. The majority of
the costs associated with the underground piping tied to the
costs of digging the trenches. The trenches must be dug
from the dispensers to the underground tanks to allow the
laying in of the vapor piping, assuring proper slope from
the dispensers down to the underground tanks . Further
costs are involved with backfilling the trenches and
repairing the pavement. Digging the trenches requires
cutting through the concrete pad over the storage tanks and
at the islands, probably shutting down the station, and
using a backhoe to dig the trench back to the underground
tanks. Costs associated with trenching are difficult to
obtain since it is not hardware related, but consists of
labor and heavy equipment charges. From a previous
analysis, EPA derived trenching and backfill costs based
upon an estimate obtained from a Stage II system installer.
This cost averaged about $30 per foot of trench.25
A great deal of importance is given to the proper
installation of the underground piping. Improper slope,
poor backfilling, and ground settling all can cause breaks
or low points in the vapor piping system. Breaks in the
vapor piping can cause vapor leaks in the system, and low
points in the piping can provide the potential for liquid
accumulation resulting in liquid blockage. Some areas of
California have indicated that as many as 50 percent of the
underground systems are incorrectly installed.26
5-11
-------
TABLE 5-4. PIPING COMPONENT DIFFERENCES BETWEEN INDIVIDUAL
AND MANIFOLDED BALANCE SYSTEM23
Number of Components
Individual Manifolded
Underground Components Balance System Balance System
Galvanized Pipe for Vapor
Risers
1" Pipe (FT) 10 10
2" Pipe (FT) 2
3" Pipe (FT) 2
3/4" Close Nipple 7 7
1" Close Nipple 13 13
2" Close Nipple 3
3" Close Nipple 6
1" Elbow 13 13
2" Elbow 6
3" Elbow 6
1" x 3/4" Reducer 7 7
2" x 1" Reducer 7
3" x 2" Reducer 3
4" x 2" Bushing 3
Fiberglass Pipe for Vapor
Return Piping
2" Pipe (FT) 476 86
3" Pipe (FT) 125
2" Threaded adapter 10 10
3" Threaded adapter 3
2" Elbow 16 2
3" Elbow 2
2" Tee 3 2
3" Tee 3
2" Coupling 9 1
3" Coupling 2
3" x 2" Reducer 4
Glued Junctions 34 26
Additional Items
4" x 3" Tank Bushing 3
2" Float Check Valve 3
Vent Manifold Drum 1
Bungs 1 1
Trenching/Assembly (ft) 165 165
5-12
-------
TABLE 5-5. TYPICAL VAPOR PIPING COSTS FOR 65,000 GALLON
PER MONTH SERVICE STATION23
Vapor Piping Costs
Individual Balance system $7,700
Manifolded Balance System $8,000
Healy Assist System $7,700
Vacuum Assist System* $7,000
8 Average of both the Hirt and Hasstech certified vacuum
assist systems.
5-13
-------
California, New York and several other Stage II areas
in the country now require tests to be conducted on the
underground piping. These tests, discussed in Chapter 6,
consist of the liquid blockage, pressure decay, and
backpressure tests. It is estimated that the costs to
perform these tests is a total of $67O.27
A common occurrence over the last several years is that
station owners across the country have been installing Stage
II underground piping whenever modifications were undertaken
that required excavation. This will reduce installation
costs for a great number of stations.
5.1.3 Affects of the UST Program
Stage II installation costs can be affected by a
simultaneous Stage II/UST program implementation by
considering the cost savings of installing Stage II at the
time underground tanks are being repaired or replaced. The
potential cost savings are realized in reduced trenching and
paving costs that would have been attributed to the Stage II
installation in the absence of any UST activity.
The key items for determining the impacts of a
simultaneous Stage II/UST program on installation costs is
to determine how many tank system leaks will occur and what
equipment will be excavated during repairs or replacement.
Several assumptions had to be made concerning the number and
type of repairs required under an UST program. These
assumptions on number or frequency of repair are drawn from
a previous analysis and are presented in Table 5-6.
Table 5-6 further summarizes the possible actions taken
in response to finding a leak in either the underground
piping or underground tank. For each remedy action, the
percent of all tank systems assumed to use that remedy is
listed. A description is added that summarizes the
resulting savings in Stage II trenching associated with each
remedy. For example, both Actions 1 and 4 (dig up all
piping, and dig up all piping and tanks) result in the
5-14
-------
TABLE 5-6. ACTIONS TAKEN IM RESPONSE TO FINDING A
LEAK IN AN UNDERGROUND TANK SYSTEM8
Percent of Costa Saved*
tn
H-i
tn
Action
Percent
of all
Systems
Description of Savings
in Stage II Piping Installation
1. Dig up all piping 4.5X
2. Dig up end of tanks only 11.8X
3. Dig up end of tanks and under 1.3X
dispensers
4. Dig up all piping and tanks 12.SX
5. Dig up only one tank 1.9X
6. Repair one leaking tank 3.1X
Total 35.OX
All trenching costs
Trenching costs over end of all tanks
Trenching costs over all tanks and under
all dispensers
AH trenching costs
Trenching costs over one tank
Trenching costs over one tank
Underground Total
Capital Capital Annual
Costs Costs Costs
65X
10X
SOX
65X
8X
SX
40X
7X
20X
40X
SX
5X
25X
5X
15X
25X
3X
3X
* Cost percentages for a typical 65,000 gallon/Month station.
-------
savings of all trenching costs. Also presented in Table 5-6
is the resulting percentage savings in total Stage II costs
that would occur under each action.
A further discussion of cost savings associated with
simultaneous Stage II/UST programs can be found in Appendix
K of the 1987 Regulatory Impact Analysis (RIA), Volume I
concerning gasoline marketing strategies.28
5.1.4 RecovervCredits
Another aspect of the annual costs for Stage II systems
is recovery credits. As discussed in Chapter 2, the return
of saturated vapors to the storage tank during vehicle
fueling eliminates the inbreathing of fresh air and
subsequent evaporation of liquid gasoline. Each gallon of
gasoline that is prevented from evaporating represents a
gallon of product the station owner can sell that would not
be present in the absence of Stage II controls. The
earnings generated from this gasoline that would have
otherwise have evaporated are counted as recovery credits.
Recovery credits may be calculated as follows.
Assuming 95 percent recovery of both displacement and
emptying losses,
recovered vapor • {(1,340 ««/HterX.95» + CC120 in5/Uter>C.95» » 1,387 as/liter.
Example of recovery credit:
1,387 us/liter x 75,700 liters x kg x liter x 12 mo. x $0.275/liter « S518/year.
MO. 10°*s 0.67kg yr
5.2 MODEL PLANT COSTS
Costs in this section are presented for the model
plants described in Chapter 2 of this report. Because of
the infinite variations in service station layout and
design, model plants were developed to represent the
Industry and to fix the physical parameters of each
facility. In addition to the items specified in Table 2-5,
such as throughput and number of nozzles, the physical
design of each model station was developed. This included
5-16
-------
distances from the dispensers to the tank to fix trenching
lengths, and designs of piping scenarios to fix piping
costs.
A detailed cost model was developed by EPA, in the 1987
Draft Regulatory Impact Analysis (RIA), that created
"ground-up" costs for each model plant.29 This model used
the piping layouts described above and detailed component
costs for aboveground equipment. Costs were obtained for
all certified equipment and averaged to estimate capital and
installed costs for each component. Costs were also
obtained for fiberglass pipe and fitting costs, installation
labor, and trenching costs. For convenience, a detailed
discussion of this model is reproduced in Appendix B of this
document.
5.3 COMPARISON OF RECENT COST STUDIES
EPA solicited and received public comments on the 1987
RIA associated with the proposal of onboard controls for
vehicle refueling. EPA received public comments concerning
Stage II costs from many sources including oil companies,
service station dealers, and auto manufacturers. Of
particular interest to EPA were comments received from the
American Petroleum Institute (API)30 and from Multinational
Business Services, Inc. (MBS)31 (under contract to the Motor
Vehicle Manufacturers Association and the Auto Importers of
America). These comments were of interest because these two
groups conducted their own cost analyses of Stage II
equipment installed in St. Louis and attempted a comparison
with the EPA cost analysis found in the Draft RIA on the
onboard proposal, (see Appendix B). The majority of the
remaining comments provided little or no cost breakdown,
making cost comparisons impossible. In addition to comments
received on Stage II costs, Pacific Environmental Services,
Inc. (PES), under EPA contract, conducted an independent
analysis of Stage II installation costs in St. Louis,
Missouri and compared the costs they obtained with the
5-17
-------
industry studies and with the Draft RIA.32 Stage II costs
in St. Louis were considered important at that time because
Stage II installations were recently completed in this
metropolitan area, and conflicting cost information was
received during the public comment period.
As stated before, the Draft RIA used a "ground-up"
model of Stage II costs, whereas, the API, MBS, and PES
studies were all surveys of Stage II costs in St. Louis. As
discussed earlier in this chapter, direct comparison of cost
surveys conducted by different groups is often difficult
especially if cost breakdowns are not available. Cost
breakdowns allow an analysis of the make-up of the costs,
and ensures that like costs are being compared (i.e., only
Stage II related costs were included in the reported costs).
Cost breakdowns and raw data for all industry surveys were
not available to allow direct comparison to EPA cost models.
5.3.1 Capital Cost Comparison
Stage II system installed capital cost estimates from
all data sources are shown in Table 5-7. These average
Stage II system costs are graphically depicted by model
plant category in Figure 5-1. This plot is useful in making
"snapshot" comparisons among the data sources for each of
the model plant categories. In order to determine a trend
or relationship among each of the subject data sets, a
linear regression method was used. The linear function was
determined as most representative, based on the use of
correlation coefficient (R-squared) values as criteria for
best fit. Figure 5-2 illustrates the relationship of
capital cost versus model plant category after the
application of the "best fit" line. No information was
presented in the API Report to explain why the "major" costs
were so much higher than the "Jobber" costs. Because of the
large differences these costs are depicted separately on
these figures.
Capital cost data submitted by API suggested that EPA
had, on average, understated costs by about 40 percent.
5-18
-------
TABLE 5-7. SUMMARY OF STAGE II SYSTEM CAPITAL COST
ESTIMATES FROM ALL SOURCES26-28'30'31
Model
Plant
No.
1
2
3
4
5
Cost Estimate Source
Draft RIA
API -Jobber
API-Major
MBS
PES
Draft RIA
API -Jobber
API -Major
MBS
PES
Draft RIA
API -Jobber
API -Major
MBS
PES
Draft RIA
API-Jobber
API -Major
MBS
PES
Draft RIA
API-Jobber
API -Major
MBS
PES
Total System
Capital Costs
$5,492
$11,262
_ »
$5,616
$5,352
$7,007
$12,168
_a
$6,517
$7,936
$11,962
$16,094
$17,479
$9,108
$12,913
$15,855
$20,020
$28,565
$11,750
$14,524
$22,917
$27,872
$41,831
$24,663
$24,523
No data reported.
5-19
-------
50,000
40,000
A
in
i
ro
o
g 30,000
20,000
10,000 -
B
A
..a
0
1
Draft RIA
.—9—
API-Jobber
Model Plant Size
MBS
API-Major
•A • •
PES
Figure 5-1. Comparison of Installed Capital Costs
Lines Based on Data Point Averages
-------
50,000
40,000
O
01
to
H
g 30,000
8
a
<3 20,000
10,000
A
0
Draft RIA
—a—
API-Jobber
A
Model Plant Size
API-Major MBS
o --*-
PES
Figure 5-2. Comparison of Installed Capital Costs
Lines Based on Linear Regression
-------
Capital costs submitted by MBS suggested EPA had, on
average, overstated costs by about 20 percent. Stage II
costs published in the Draft RIA with the onboard proposal
fell between the costs submitted by these commenters. In
addition, the St. Louis data obtained by PES also fell
between the API and MBS costs and compared favorably (within
5 percent) with the Draft RIA costs. The fitted curves of
Figure 5-2 illustrate that PES1 costs were close to the
Draft RIA costs for the smaller model plants and between the
Draft RIA and API costs for the larger model plants.
5.3.2 Annual Cosj; Comparison
The commenters supplied annual costs associated with
the estimated capital costs of the Stage II systems on a
model plant basis. However, difficulties arose in
summarizing and comparing these costs because each commenter
used different cost assumptions for: (1) annualized cost of
capital, (2) maintenance costs, (3) recovery credits, and
(4) the number of nozzles assigned to each model plant. In
an effort to normalize these variations, each capital cost
estimate presented in Section 5.3.1 was converted to
annualized costs using EPA's cost methodology from the Draft
RIA and using the same assumptions for equipment life
(8 years aboveground equipment, 35 years below-ground
equipment), interest rate (10 percent), taxes and insurance
(4 percent), and calculation and costs dealing with recovery
credits.33 Maintenance costs were considered the same for
each annual cost estimate.
Table 5-8 summarizes the annual cost estimates
normalized using the assumptions above. These estimates are
graphically depicted in Figures 5-3 and 5-4.
5.4 CURRENT COSTS OP STAGE II SYSTEMS
Based on the comparisons discussed in Section 5.3, it
can be concluded that the ground-up model from the Draft RIA
(reproduced and presented in Appendix B) provided a
reasonable estimate of actual Stage II installations. This
5-22
-------
TABLE 5-8. SUMMARY OF NORMALIZED STAGE II SYSTEM ANNUAL
COST ESTIMATES FROM ALL SOURCES
Model
Plant
No.
1
2
3
4
5
Normalized Annual
Cost Estimate Source Costs
Draft RIA
API -Jobber
API -Major
MBS
PES
Draft RIA
API -Jobber
API -Major
MBS
PES
Draft RIA
API -Jobber
API-Major
MBS
PES
Draft RIA
API -Jobber
API -Major
MBS
PES
Draft RIA
API -Jobber
API-Major
MBS
PES
$1,270
$2,045
NAa
$1,288
$1,244
$1,280
$1,953
NAa
$1,195
$1,515
$2,380
$2,848
$3,163
$1,893
$2,559
$2,960
$3,363
$4,764
$2,230
$2,726
$2,430
$2,833
$5,129
$2,765
$2,847
Cannot be calculated since no capital costs reported.
5-23
-------
1
01
I
to
6,000
5,000
4,000
3,000
2,000
1,000
A
A
I
Draft RIA
0
API-Jobber
-0
Model Plant Size
API-Major MBS
..A. --•-
PES
A-
Figure 5-3. Comparison of Annual Costs
Lines Based on Data Point Averages
-------
6,000
5,000
O
o
8
I
"
Ul
K)
Ul
4,000
3,000
2,000
1,000
n
o
Draft RIA
B
API-Jobber
A
Model Plant Size
API-Major MBS
aL._
"r-
PES
Figure 5-4. Comparison of Normalized Annual Costs
Lines Based on Linear Regression
-------
model was, therefore, used to estimate current 1991 Stage II
costs. The model in Appendix B was used, but replacing key
component costs to reflect 1991. Table 5-9 contains a
summary of the cost data changed from the Draft RIA analysis
to generate 1991 costs. As stated earlier in this document,
multi-product dispensers, offering each of three gasoline
grades on each side of the dispenser, have increased in
popularity in recent years. The Draft RIA made an attempt
to estimate the mix of single and multi-product dispensers
to calculate a national Stage II cost impact. For purposes
of this document two separate estimates have been made, one
to represent single dispensers and one to represent multi-
product dispensers. Table 5-10 summaries 1991 capital costs
of Stage II systems for single dispenser facilities and 1991
capital costs for multi-product dispensers. Annualized
costs were also calculated using the approach discussed in
Appendix B, but using the 1991 capital costs and the 1991
RVP and gasoline price for recovery credit calculations.
Table 5-11 summarizes annual costs for single and multi-
product dispensers, respectively.
Another important factor to consider when reviewing
Stage II costs is system cost effectiveness. Cost
effectiveness is the annual operating costs divided by the
annual emission reduction, yielding a value of dollars spent
per unit measure of emission reduction. Table 5-12 presents
the 1991 cost effectiveness of Stage II systems, expressed
as dollars per megagram of emission reduction. Again,,
values are presented for both single and multi-product
dispensers facilities.
The program effectiveness or overall emission reduction
is dependent on the exemption level selected, as indicated
in Section 4.4.3. The cost effectiveness of the program is
also dependent on the exemption level imposed. Smaller
facilities have higher cost effectiveness values (see Table
5-12). Program cost effectiveness, therefore, improves by
5-26
-------
TABLE 5-9. SUMMARY OF COST ITEMS CHANGED IN APPENDIX B
COST MODEL TO OBTAIN 1991 COSTS
Item Cost
Nozzle Costs (New)
Emco Wheaton 236.84
OPW 221.05
Husky 237.60
Nozzle (Rebuilt)
Emco Wheaton 192.98
EZ-flo 144.74
(OPW or Emco Wheaton)
Component Costs (Spout kit)
Emco Wheaton 26.56
Husky 20.86
OPW 17.46
EZ-flo (OPW or Emco wheaton)
Boot Kit
EZ-flo 22.26
Husky 50.65
Emco Wheaton 35.78
Hoses (10 ft, w venturi)
Thermoid 237.50
Goodyear 246.36
Dayco 389.54
Hoses (10 ft., w/o venturi)
Thermoid 141.94
Goodyear 155.87
Dayco 125.16
Breakaways (one time)
Dayco 47.70
Husky 66.65
Breakaway (reconnectable)
Husky 143.30
Petroleum 180.31
EMCO Wheaton 125.35
5-27
-------
TABLE 5-9. SUMMARY OF COST ITEMS CHANGED IN APPENDIX B
COST MODEL TO OBTAIN 1991 COSTS (CONTINUED)
Item Cost
Miscellaneous Equipment
12" whiphose
Goodyear 42.54
Thennoid 48.76
Dayco 47.69
Retractor Clamp
Goodyear 10.26
Thermoid 9.06
EZ-flow (Dayco) 6.45
(Goodyear, Thermo, and 7.17
Gates)
High Hang Hose Retractors
Catlow 163.00
96.30
Swivels 50.50
5-28
-------
TABLE 5-10. 1991 STAGE II BALANCE SYSTEM CAPITAL COST
COMPONENT
COST OF COMPONENT
MODEL PLANT 1
Number of Nozzles
Dispenser Direct Cost
Piping Direct Cost
Total Capital Cost
SINGLE
DISPENSER
2
1,580
3f910
5,490
MULTI PRODUCT
DISPENSER
4
3,210
3,910
7,120
MODEL PLANT 2
Number of Nozzles
Dispenser Direct Cost
Piping Direct Cost
Total Capital Cost
MODEL PLANT 3
Number of Nozzles
Dispenser Direct Cost
Piping Direct Cost
Total Capital Cost
MODEL PLANT 4
Number of Nozzles
Dispenser Direct Cost
Piping Direct Cost
Total Capital Cost
MODEL PLANT 5
Number of Nozzles
Dispenser Direct Cost
Piping Direct Cost
Total Capital Cost
3
2,370
4,950
7,320
6
4,740
7,860
12,600
9
7,120
9,690
16,810
15
11,860
12,650
24,510
6
4,810
4,950
9,760
12
9,620
7,860
17,480
18
14,430
9,690
24,120
30
24,060
12,650
36,710
5-29
-------
TABLE 5-11. 1991 STAGE II BALANCE SYSTEM ANNUAL COST
COMPONENT
MODEL PLANT 1
Capital Recovery Cost
Maintenance Cost
other Indirect Costs
Recovery Credit
Total Annual ized
Cost
MODEL PLANT 2
Capital Recovery Cost
Maintenance cost
Other Indirect Costs
Recovery Credit
Total Annual ized
Cost
MODEL PLANT 3
Capital Recovery Cost
Maintenance Cost
Other Indirect Costs
Recovery Credit
Total Annual ized
Cost
MODEL PLANT 4
Capital Recovery Cost
Maintenance Cost
Other Indirect Costs
Recovery Credit
Total Annual ized
Cost
MODEL PLANT 5
Capital Recovery Cost
Maintenance Cost
Other Indirect Costs
Recovery Credit
Total Annual ized
Cost
COST OF
SINGLE
DISPENSER
701
475
219
129
1,266
939
617
293
518
1,331
1,668
1,230
504
906
2,496
2,297
1,852
672
1,683
3,138
3,455
3,090
980
4,790
2,735
COMPONENT
MULTI PRODUCT
DISPENSER
893
475
285
129
1,524
1,555
617
485
518
2,139
2,313
1,230
699
906
3,336
3,298
1,852
965
1,683
4,432
5,175
3,090
1,468
4,790
4,943
5-30
-------
TABLE 5-12. COST EFFECTIVENESS OF 1991 STAGE II
BALANCE SYSTEMS8
Single Multiproduct
Dispenser Dispenser
MODEL PLANT 1
Annuali zed Costs, $
Emission Reduction,
Cost Effectiveness,
MODEL PLANT 2
Annualized Costs, $
Emission Reduction,
Cost Effectiveness,
MODEL PLANT 3
Annualized Costs, $
Emission Reduction,
Cost Effectiveness,
MODEL PLANT 4
Annualized Costs, $
Emission Reduction,
Cost Effectiveness,
MODEL PLANT 5
Annualized Costs, $
Emission Reduction,
Cost Effectiveness,
Mg
$/Mg
Mg
$/Mg
Mg
$/Mg
Mg
$/Mg
Mg
$/Mg
1,266
0.34
3,680
1,331
1.0
1,290
2,496
1.8
1,380
3,138
3.4
910
2,735
9.7
280
1,524
0.34
4,430
2,139
1.0
2,070
3,336
1.8
1,850
4,432
3.4
1,290
4,943
9.7
510
* Emission reduction from Table 3-8, and assuming annual
enforcement (86 percent in-use efficiency).
5-31
-------
exempting higher cost facilities. Table 5-13 summarizes
program cost effectiveness when compared to certain
exemption levels. This table was calculated based upon the
model plant cost effectiveness values presented in Table
5-12 and the model plant distribution values contained in
Tables 2-8 and 2-10. Values are presented for facilities
with either single dispensers or multiproduct dispensers, as
in Table 5-12, but also an average cost that assumes equal
distribution of single and multiproduct dispensers.
5-32
-------
TABLE 5-13. PROGRAM COST EFFECTIVENESS COMPARED
TO EXEMPTION LEVEL
Program
Exemption Level
Program Cost Effectiveness
($/Mg)
Single Multiproduct
Dispenser Dispenser Average'
No Exemptions
1,130
1,570
1,350
Ex < 2,000 gal/month
1,030
1,460
1,240
Ex < 10,000 gal/month
890
1,310
1,100
Ex < 10,000 gal/month
Independents
< 50,000 gal/month
820
1,210
1,020
Average assumes equal distribution of single and
multiproduct dispensers.
5-33
-------
5.5 REFERENCES
1. Memorandum from Norton, R., Pacific Environmental
Services, Inc. (PES), to Shedd, S., U.S.
Environmental Protection Agency. April 29, 1991.
Trip Report - New York Department of Environmental
Conservation.
2. Telecon. Bowen, Elizabeth, Pacific Environmental
Services, Inc. (PES) with Pratt, B., State of
Missouri. May 30, 1991. Stage II Program
Implementation.
3. EMCO Wheaton Price List, March 1991.
4. Telecon, Bowen, E., Pacific Environmental
Services, Inc. (PES) with Holcom, C., Husky
Corporation. April 30, 1991. Stage II Nozzle
Costs.
5. Telecon Bowen, E., Pacific Environmental Services,
Inc. (PES), with Taylor, B., OPW. May 29, 1991.
Stage II Nozzle Costs.
6. Reference 3.
7. Reference 3.
8. Reference 4.
9. Draft Regulatory Impact Analysis: Proposed
Refueling Emission Regulations for Gasoline-Fueled
Motor Vehicles — Volume I - Analysis of Gasoline
Marketing Regulatory Strategies. U.S.
Environmental Protection Agency, Office of Air
Quality Planning and standards an and Office of
Mobile Sources. Publication No. EPA-450/3-87-
OOla. July 1987. Appendix B, Table B-15.
10. Telecon. Norwood, P., Pacific Environmental
Services, Inc., with Friedman, G., EZ-flo.
October 4, 1991. Stage II Equipment Costs.
11. Reference 3.
12. Fax communication to Bowen, E., Pacific
Environmental Services, Inc. (PES), from Carmack,
M., Catlow. May 8, 1991. Price List for Stage II
Equipment.
13. Reference 9, Appendix B, Table B-2.
5-34
-------
14. Telecon. Bowen, E., Pacific Environmental
Services, Inc. (PES) with Terlizzi, L., Thermoid.
April 30, 1991. Stage II Equipment Costs.
15. Telecon. Bowen, E., Pacific Environmental
Services, Inc. (PES) with Whittington, G.,
Goodyear. May 6, 1991. Stage II Equipment Costs.
16. Fax Communication to Bowen, E., Pacific
Environmental Services, Inc. (PES) from Gelle, B.,
Dayco. May 15, 1991. Stage II Equipment Price
List.
17. Reference 12.
18. Reference 14, 15, 16.
19. Reference 13.
20. Reference 9, Appendix B, Table B-5.
21. Reference 20.
22. Reference 20.
23. Reference 20.
24. Reference 9, Appendix B, Table B-ll.
25. Reference 24.
26. Memorandum from Norwood, P., Pacific Environmental
Services, Inc. (PES) to Shedd, S., U.S.
Environmental Protection Agency. April 30, 1991.
Trip Report to California Agencies to Discuss
Stage II Programs.
27. Wakim, Paul, C. J. Sample, K.A. Rooney, and D.
Clemons (American Petroleum Institute). API
Survey of Actual Stage II Implementation Costs In
The St. Louis Metropolitan Area. American
Petroleum Institute. December 2, 1988.
28. Reference 9, Appendix K.
29. Reference 9, Appendix B.
30. Reference 26.
31. Responses to Points Raised by EPA Concerning the
MBS Study "Costs and Cost-Effectiveness of Stage
II and Onboard Refueling Vapor Controls" (April
1987). Multinational Business Services, Inc.
(MBS). October 1987.
5-35
-------
32. Norton, R. and Scott Osbourn (Pacific
Environmental Services, Inc.)- Evaluation of
Stage II Vehicle Refueling Control Equipment
Installation Costs. Prepared for U.S.
Environmental Protection Agency. November 28,
1988.
33. Reference 9.
5-36
-------
6.0 PROGRAM IMPLEMENTATION
As discussed in Chapter 1, Stage II vapor recovery has
been a part of VOC emission control in California for some
time. Since the introduction of Stage II vapor recovery
California in the early 70's, this program has become one of
California's major VOC control strategies. Seventeen
districts in California containing areas that are classified
nonattainment for ozone have Stage II programs that have
been in effect for over a decade. The remaining districts
in California have also recently adopted regulations
requiring Stage II vapor recovery for benzene control.
Other areas of the country have also established Stage
II vapor recovery programs. The District of Columbia
implemented a Stage II program in the early 19SOs and
Missouri adopted vapor recovery regulations in the St. Louis
area later in the 80s. In the late 1980s and early 1990s
several other States and local agencies have adopted Stage
II programs. These areas include New Jersey, New York (New
York City metropolitan area) Massachusetts, Pennsylvania,
Washington, Oregon, and Dade County, Florida. The CAAA of
1990 require the installation of Stage II vapor recovery
systems in many ozone nonattainment areas. Based on final
nonattainment designations, this would affect almost 60
metropolitan areas in the United States.
The purpose of this chapter is to provide information
on the planning, implementation, and enforcement of Stage II
programs in other States. Incorporated into this discussion
are examples of how areas with current Stage II programs
handle certain situations and issues. This ranges from
experience in areas such as San Diego which has almost 20
years experience with Stage II to areas such as
6-1
-------
Massachusetts and Dade County, Florida with programs only
recently adopted. Appendix F provides summaries of many of
the programs in the United States. For each program,
Appendix F provides a description of the program with
problems encountered and recommendations for new areas based
on their experience. In addition, items such as permit
applications, inspection checklists, etc. are included for
some of the areas in Appendices G-K. Specifically, this
chapter addresses planning elements, regulations, and
permitting and enforcement considerations. The 1PA
enforcement guidance document should be consulted for
guidance on enforcement issues.
6.1 PLANNING
The planning of a Stage II program involves several
considerations including the characterization of the
affected industry and the estimation of environmental and
economic impacts. The information contained in other-
chapters of this document can aid in the determination of
some of these factors.
An important consideration from the outset of Stage II
program planning is to work closely with other agencies that
may be affected by the program. For instance, the
department or agency responsible for the measurement and
accuracy aspects of gasoline dispensers would probably have
an interest in such a program. Other agencies that are
concerned with safety aspects, such as the Occupational
Safety and Health Administration (OSHA) and the Fire
Marshal, will also be affected by Stage II and should be
consulted. The significance of working with these types of
agencies is evident in the California certification process
discussed in Chapter 4. Before a Stage II system is
certified, it must meet the approval of California Division
of Measurement Standards, California OSHA, and the
California Fire Marshal, in addition to meeting the
requirements of the California Air Resources Board (CARB).
6-2
-------
It may be beneficial to contact these types of agencies at
the beginning and solicit their involvement with the Stage
II program.
6.1.1 Characterization of the Affected Industry
Chapter 2 characterizes the industry affected by Stage
II regulations. A service station is defined as any site
where gasoline is dispensed to motor vehicle fuel tanks from
stationary storage vessels. This includes public (retail)
and private facilities. Miscellaneous retail outlets that
are considered service stations include conventional service
stations, convenience stores, mass merchandisers, marinas,
parking garages, and other similar facilities which sell
gasoline to the public. Private facilities include those
locations where gasoline is dispensed into government agency
(Federal, military, State, and local) vehicles, fleet (auto
rental, utility companies, taxis, school buses, etc.)
vehicles, and trucking and local service vehicles.
In order to estimate the impacts of a Stage II
regulation, it is necessary to identify the number of
facilities potentially affected and the volume of gasoline
dispensed at these facilities.
6.1.1.1 Number of Facilities. The number of
facilities can be estimated using a variety of techniques.
Since most areas that will be required to install Stage II
have previously been classified as nonattainment for ozone,
it is likely that Stage I vapor recovery regulations exist
in these areas. The Stage I permit files can be used to
supply an estimate of the number of potentially affected
facilities. Other possible sources of this type of
information are records pertaining to underground storage
tanks, Department of Weights and Measures, tax records,
local fire departments or even phone directories.
In the absence of actual records or data, local or
State trade organizations could be contacted. Also,
information such as the survey completed by NPN discussed in
Chapter 2 provides retail service station numbers on a State
6-3
-------
basis. These could be used and adjusted to a smaller
geographic area using a factor such as population or
gasoline throughput.
6.1.1.2 Area Gasoline Throughput. The combination of
the area gasoline throughput and the emission factors
discussed in Chapter 3 will provide an estimate of the
uncontrolled emissions from vehicle refueling. If gasoline
taxes are imposed in the study area, records relating to
gasoline sales should be available at the tax office. If
the study area entails an entire State, NPN annually
estimates gasoline consumption on a State basis. Gasoline
consumption and methods of estimating gasoline consumption
on a county level are also discussed in Chapter 2.
6.1.1.3 Size Distribution of Facilities. The
distribution of facilities by throughput and according to
the number of nozzles is important. Ideally, an agency
could obtain detailed information regarding the number of
service stations, the associated gasoline throughput, and
the number of nozzles. However, in the absence of the
resources necessary to develop such a database, it is
possible to draw comparisons between the areas covered by
the Lundberg data discussed in Chapter 2 and summarized in
Appendix A and the agency's regulated area, The data can be
used to estimate size distributions for counties in
designated population ranges or with a known number of
service stations. For example, if a county's population is
approximately 50,000, the counties of Union, Hudson, and
Monmouth, New Jersey could be selected from Appendix A as
counties with comparable populations. The size
distributions of these three counties could then be averaged
to predict a size distribution for the study area county.
Model plants could then be developed which include the
number of nozzles and gasoline throughput. Alternatively,
the model plants provided in Chapter 2 may be used. The
distribution of facilities could be applied to the model
6-4
-------
plants to estimate the number of facilities represented by
each model plant.
6-1.2 Estimation of Impacts
The population and distribution of facilities, gasoline
consumption, individual facility costs, and planned
enforcement levels are used to predict environmental and
economic impacts.
6.1.2.1 Environmental Impacts. The emission
reductions anticipated from the regulation may be estimated
by calculating the uncontrolled emissions and multiplying
these emissions by the expected overall effectiveness for
the program. The uncontrolled emissions can be calculated
by multiplying the gasoline throughput by the uncontrolled
emission factor discussed in Chapter 3. The overall, or in-
use, effectiveness may be estimated according to the
expected level of effort which the agency plans to have
available for the program. In-use effectiveness is
discussed in detail in Chapter 4.
In order to evaluate the impacts associated with
exemption levels, the throughput for the number of
facilities in model plants that fall below the anticipated
exemption cutoff should not be multiplied by the selected
control level or use the Stage II program efficiencies shown
in Chapter 4 with exemption levels already assumed.
6.1.2.2 Economic Impacts. Costs initially must be
estimated on a facility basis. The agency may choose to
gather information specific to their area regarding
installation, equipment, and maintenance costs for these
systems. If resources are not available for such a detailed
analysis, Chapter 5 discussed costs of Stage II with model
plant costs. Model plant costs may then be multiplied by
the number of facilities assigned for each model plant to
estimate the total area impacts.
The overall cost in relation to the emission reduction,
or cost effectiveness, may then be calculated by dividing
the overall cost by the overall emission reduction. Since
6-5
-------
the cost effectiveness for smaller facilities is higher due
to the lower gasoline throughput and resulting lower
emission reduction and recovery credit, cost effectiveness
is often used to define exemption levels for these smaller
facilities.
6.1.3 Public Awareness
Public acceptance is vital to the success of any Stage
II program. The slight variations in the operation of Stage
II equipment can annoy uninformed customers and lead to
improper use possibly reducing efficiency and the incorrect
conclusion that the equipment is faulty. Therefore, an
agency should consider ways to inform and educate the public
about the Stage II program. Many regulations require that
operating instructions be placed at the pump. This is
perhaps the simplest and most straightforward method of
providing the public information about the operation of
Stage II equipment.
Another method used, especially in California, is a
toll free complaint number. The number is placed on the
pump with the operating instructions and is specifically for
Stage II complaints. California officials have indicated
that in the earlier periods of Stage II, these lines were
used by the public often to express discontent with Stage
II. However, as the public has become more aware of the
equipment, the complaint lines have evolved into a form of
public compliance program, where persons call in with
reports of faulty or missing equipment.
In addition to the operating instructions and telephone
number, the agency can develop a public awareness program.
The publication and distribution of brochures, pamphlets,
fact sheets, etc. is a manner of providing information to
the public. Such a pamphlet from Massachusetts is provided
in Appendix G-l. The use of the media to describe Stage II
has been used successfully in California. Television,
radio, and newspaper spots have described the environmental
6-6
-------
and personal health benefits associated with Stage II and an
explanation of operating procedures.
While these public awareness measures are important to
gain acceptance of Stage II, service station employee
awareness and education may have a more significant impact
on reducing emissions. It is extremely helpful if these
employees are knowledgeable of the operation and maintenance
requirements of Stage II equipment. There are several ways
that an agency can promote this. They can provide
workshops, training courses, etc. for service station
employees that discuss Stage II equipment, regulations, and
inspection procedures. The agency could also promote self-
inspection programs that encourage station employees to
conduct periodic equipment inspections to ensure that the
equipment is in proper condition. Appendix G-2 contains a
self inspection handbook published by the California Air
Resources Board that is provided to station owners. An
informed and conscientious service station employee
population will decrease the enforcement effort needed and
the excess emissions from vehicle refueling.
6.2 REGULATIONS
Development of appropriate rules is necessary in order
to satisfy the intent of the program and determine
individual facility compliance. As with any regulation,
Stage II regulations should be clearly written and specific.
The rules should contain definitions; requirements for the
equipment installation, operation, and maintenance;
exemptions levels; compliance schedules; and testing and
recordkeeping requirements. Many Stage II regulations also
require that operating instructions be posted at the pumps.
Copies of many current Stage II regulations are contained in
Appendix H.
6.2.1 Equipment Requirements
Most current Stage II regulations contain a statement
that prohibits gasoline refueling without a certified or
6-7
-------
approved Stage II system. Common language for this
requirement is "No owner or operator shall transfer, permit
the transfer, or provide equipment for the transfer of
gasoline from a stationary storage tank at a service station
into a motor vehicle fuel tank unless an approved Stage II
vapor recovery system with 95 percent or greater efficiency
is installed and used during the transfer."
This language brings to light an important point, the
definition of an "approved Stage II vapor recovery system."
An "approved Stage II vapor recovery system" is defined in
various ways but in all current situations is directly or
indirectly linked to certification by the California Air
Resources Board that the system controls VOC emissions with
95 percent efficiency. In California, an approved system is
any CARB certified system. CARB certification and Executive
Orders are discussed in Chapter 4. In addition, Appendix C
contains the certification testing procedures and Appendix D
addresses Executive Orders. Most states and local agencies
automatically approve, or certify, Stage II systems that
have been certified by CARB. EPA is not aware of any State
or local agency that has conducted testing and certified
Stage II equipment which has not been previously CARB
certified. However, most regulations outside of California
do allow the possibility of non-CARB certification, although
no specific test methods or procedures are identified.
While the universe of certified equipment in non-
California areas has not been broadened to include equipment
not CARB certified, many areas are limiting the approved
equipment from the complete list that is currently certified
by CARB. For instance, both Massachusetts1 and Dade County,
Florida2 allow only coaxial hoses. Dade County permits only
the most recent generation of nozzles and other equipment.
These are options available to a beginning program that can
reduce the confusion as to what is "approved", as well as
ensuring use of the prevailing technology. In fact, CARB
6-8
-------
representatives have indicated that they feel this is a
sound approach for new programs.3
In all circumstances, it is important that both
industry and inspectors be completely aware of those systems
and equipment which are approved and acceptable for an area.
Even if an agency accepts GARB certification to determine
approvable systems, it can maintain an up-to-date listing
available to all parties that clearly specifies the
permissible equipment and combinations of components. This
is generally the approach being taken by the New York State
agency.4
6.2.2 Exemption Levels
The CAAA of 1990 require that gasoline dispensing
facilities with more than 10,000 gallons of gasoline
throughput per month (50,000 gallons per month in the case
of an independent small business marketer) install Stage II.
Therefore, by legislative mandate, the maximum exemption
levels which a State or local agency may adopt are clearly
defined. However, there are several variations that may be
incorporated.
Due to the difficulty of determining the stations that
fall under the definition of "independent small business
marketer", many areas choose not to have a separate
exemption level for this group. This is allowed under the
Clean Air Act, as discussed in Chapter 1. In fact,
presently no agency exempts independent marketers at a
different throughput level from the remainder of the service
station population. Many areas choose not to have any
exemption level at all and require that all gasoline
dispensing facilities install Stage II equipment.
Pennsylvania's Stage II regulations contain an
additional exemption requirement. Initially, all stations
with monthly throughputs of 10,000 gallons per month or more
are required to install Stage II equipment. In addition,
whenever a station, regardless of throughput, is constructed
or modified it is required that Stage II equipment be
6-9
-------
installed. Massachusetts1 regulations also contain similar
requirements. This eliminates a large portion of the
installation cost and lessens the impacts on smaller
stations.
It is important that the regulation include specific
stipulations and procedures to verify exemption status. As
the CAAA specify exemptions based on gasoline sales, or
throughput, it is anticipated that most regulatory agencies
will follow this example, although Missouri's Stage II
regulations contain an exemption level related to storage
tank capacity (2,000 gallons for agricultural usage).
Agencies with Stage II vapor recovery programs have
indicated that problems exist with the verification of
facility throughput and, thus, the identification of exempt
facilities. One approach is to shift the burden of proof
from the agency to the facility. The Bay Area Air Quality
Management District (Bay Area) regulations make it apparent
that the burden of proof lies with the facility. The
regulation states that "the burden of proof of eligibility
for exemption from this rule is on the applicant. Persons
seeking such an exemption shall maintain adequate records
and furnish them to the Air Pollution Control Officer (APCO)
upon request." This allows the agency to evaluate not only
the throughput data but the adequacy of the data provided.
This situation can also be avoided by specifying
procedures for keeping records and determining throughput.
For instance, New York's regulation states, "The sum of all
gasoline deliveries to a gasoline dispensing site during the
previous 12 consecutive months will be used to determine
whether the requirements of section 230.2 of this Part
apply. Once a gasoline-dispensing site becomes subject to
the requirements of section 230.2 because its annual
gasoline throughput exceeds an applicability level,
subsequent decreases in gasoline deliveries or throughput do
not excuse a source owner from having to maintain the
effectiveness of the stage I and/or stage II equipment."
6-10
-------
6.2.3 Compliance Schedules
The CAAA of 1990 contain specific provisions related to
compliance dates. Section 182(e)(3) states that within 2
years from the enactment of the CAAA of 1990, States must
"submit a revision to the applicable implementation plan to
require all owners or operators of gasoline dispensing
systems to install and operate ... a system for gasoline
vapor recovery of emissions from the fueling of motor
vehicles." It also designates compliance dates as follows:
(i) 6 months after the adoption date, in the
case of gasoline dispensing facilities for which
construction commenced after the date of the
enactment of the Clean Air Act Amendments of 1990;
(ii) one year after the adoption date, in
the case of gasoline dispensing facilities which
dispense at least 100,000 gallons of gasoline per
month, based on average monthly sales for the 2-
year period before the adoption date; or
(iii) 2 years after the adoption date, in the
case of all other gasoline dispensing facilities.
Any gasoline dispensing facility described under
both clause (i) and clause (ii) shall meet the
requirements of clause (i).
The determination of an appropriate and realistic
compliance schedule within the CAAA requirements involves
the study of many factors. The schedule for installation of
Stage II equipment should allow sufficient time for
facilities to plan for their needs, as well as alleviating
any contractor shortages and potential premium charges. In
most instances, the compliance schedule is multi-phase, with
facilities with larger gasoline throughputs required to
install the Stage II equipment in the initial phase and the
smaller stations following. This originally would affect
the larger oil companies and jobbers, and help to avoid
competition between these facilities and smaller businesses
for contractors. This method also affects a larger
percentage of the gasoline throughput in the shortest time
frame. Under Section 325 of the CAAA, of 1977 a three year
6-11
-------
phase-in for independent small business marketers is
provided.
In determining whether a compliance schedule is
reasonable, the major issues to investigate are: (l) the
number of contractors in an area; (2) the number of service
stations in each cutoff classification; and (3) the
equipment availability due to other areas in the region or
country that are simultaneously requiring the installation
of Stage II systems. Table 6-1 summarizes the exemption
levels and compliance schedules of various Stage II
programs.
6.2.4 Recordkeepina Requirements
The most common recordkeeping requirement pertains to
gasoline sales or throughput. In many instances, throughput
is determined by keeping records on the amount of gasoline
delivered to the site, although the CAAA of 1990 specify
exemptions based on gasoline sales. It is appropriate that
records be kept for either, or both, deliveries and sales.
An additional check of gasoline sales could be obtained from
tax records, or the facility could be required to obtain and
keep this tax information on-site along with the facility
generated data. It is also possible that recordkeeping
requirements could be added as permit conditions. Some
areas have a recordkeeping requirement that results of
installation tests be kept on site. These tests are
discussed in detail in Section 6.3.3.
€.3 PERMITTING
Permits are a tool that local air pollution control
agencies can use in getting Stage II vapor recovery control
systems installed properly. The permits and permit
conditions should be clearly written to avoid confusion on
the part of the owner/operator of the facility and to
enhance enforcement efforts. Several aspects of permitting
are discussed in more detail in the following sections,
including the identification of sources, permit forms and
6-12
-------
TABLE 6-1. SUHMARY OF STAGE It PROGRAM EXEMPTION LEVELS AND COMPLIANCE SCHEDULES
(As of June 1991)
State/Regulatory Agency
Covered Area
Exemption Levels
Compliance Schedule
California*
Bay Area AOMD
San Francisco area
cr>
South Coast MUD
San Diego APCO
Lot Angeles area
San Diego area
Storage tanks with capacity < 260
gal. and used for "Implements of
husbandry"
Miere the District determine* that
Stage II is not feasible
Vehicle to vehicle refueling
Facilities that exclusively fuel
motor vehicle tanks < 5 gallons
Facilities that exclusively fuel
aircraft
Facilities with < 60,000 per year
throughput where Stage II us* not
installed before July 1, 1983
Facilities with 75 percent of
throughput for fueling Implements of
husbandry
Retail stations with storage tanks
less than 260 gallons
Nonretail stations with storage
tanks less than 550 gallons
Nonretail stations with less than
2,000 gallon per month throughput
for the facility
Dispensing fro* any tntemediate
refueler
Dispensing of natural gas or propane
Hhen not nixed with another VOC
Into vehicles performing emergency
work
Storage tanks used primarily for the
fueling of aircraft or boats
The Bay Area District has
had Stage II requirements
since the 1970s
The South Coast District
has had Stage II
requirement* since the
1970s
The San Diego District has
had Stag* II requirement*
since the 1970s
-------
TABLE 6-1. SUMMARY OF STAGE II PROGRAM EXEMPTION LEVELS AM COMPLIANCE SCHEDULES
CONTINUED)
State/Regulatory Agency
Covered Area
Exemption Level*
Compliance Schedule
District of Columbia
Missouri
Washington, B.C.
St. Louis area
01
I
Hen Jersey DEP
Entire State
New York DEC
New York City area
All dispensing facilities available
to the general public by virtue of
having Military status having 3 or
less dispensing nozzles
Stationary storage tanks having •
capacity < 2,000 gallons and used
for fueling "Implements of
husbandry"
Stationary storage tanks having a
capacity < 2,000 gallons installed
before September 15, 1976
< 10,000 gallons/month
Dispensing device* at a marina used
exclusively for marine vehicles
Site specific determination that
Stage II is technically or
economically Infeasible
In accordance with the DC
Air Pollution Control Act
of 1984
Final compliance date for
all source* was December
31. 1987.
30, 1988 for
facilities > 40,000
gal/month and December 29,
1989 for facilities >
10,000 gal/month
July 1, 1988 for
facilities > 500,000
gal/year and July 1, 1989
for facilities > 250,000
gal/year
-------
TABLE 6*1. SUMMARY OF STAGE 11 PROGRAM EXEMPTION LEVELS AND COMPLIANCE SCHEDULES
(CONTINUED)
State/Regulatory Agency
Covered Area
Exemption Levels
Compliance Schedule
Massachusetts DEN
Entire State
< 20,000 gal/month constructed or
modified before November 1, 1989
Florida/ftade County DEP
Ml Ml area (Dade County)
Pennsylvania DEM
Philadelphia area
tn
Marinas servicing boats
Airports servicing airplanes
Established stations < 10,000
gal/month
< 10,000 gal/month constructed
modified before June 25, 1990
April 1, 1991 for
facilities > 1,000,000
gal/year; April 1. 1991
for facilities > 500.000
gal/year; end April 1,
1993 for facilities >
20,000 gal/Mnth
I Mediately for nstt
facilities and December
14, 1992 for existing
facilities
June 25, 1991 for
facilities > 1,500,000
gal/year; December 25,
1991 for facilities >
1,000.000 gal/year; June
25, 1992 for facilities >
500,000 oaI/year; and June
25, 199S for facilities >
10,000 gel/month
All local district* In California have responsibility for Stage II (Phase II) programs. The ley Art*, South Coast, and San Diego districts
shown in the Table and 14 other districts that are nonet tal patent for ozone have had Stage II regulations for aver • decade. The remaining
districts required Stage II be installed for bensene control by 1991. While the model regulation provided by CAW (see Appendix F.1)
suggested a throughput cutoff of 480,000 gallons per year (40,000 gallons per month), the Districts Implemented a variety of cutoffs ranging
from no exemptions to this 480,000 gal/year level. These Districts are discussed in Appendix E as they are those with the most experience
with Stage II.
-------
applications, the issuance of operating permits, and testing
requirements. Appendix I contains information related to
permitting.
6.3.1 Identification of Sources
While estimates of the number of facilities may be
obtained from a variety of sources as discussed in Section
6.1, the actual identification of sources to be contacted
for permitting purposes can be difficult. An analysis of
the methods used for this identification process by agencies
with the newest Stage II programs reveals several
approaches.
Stage I permit records can be of great assistance in
this identification. New Jersey5 and Dade County, Florida6
relied on these files. New Jersey sent a letter to all
facilities in the Stage I permit system and informed them
that they were required to obtain a Stage II permit and
install the equipment. Dade County also used information
from their underground storage tank permitting program to
complement the Stage I data.
Pennsylvania identified sources by contacting major oil
companies and obtaining information from the State
Department of Licensing and Inspection.7 Massachusetts used
tax records to identify sources. Each source was then sent
a Registration and Classification form which was returned to
the Agency, who contacted the facilities which needed a
permit.8
6.3.2 Permit Forms and Applications
The permit form and application is the best means of
obtaining information regarding a facility and the type of
equipment to be installed. The forms should be designed to
allow the department to easily obtain the important
information without requiring a great deal of excess data.
An obvious requirement for the permit application is the
name and address of the facility. However, in addition to
this information it is beneficial to include the name and
address of the business owner, the operator/lessee, and a
6-16
-------
site contact. The nature and purpose of the application
should be stated. Station characteristics such as the
operating schedule, monthly and annual throughput, and
number of nozzles, hoses, and dispensers should be provided.
Information pertaining to the type of Stage II system to be
installed should also be included. Specifically, this
should consist of the equipment to be installed; a
preliminary site plan of all tanks, dispensers, and
underground piping. Most current Stage II permit forms
require that the CARB Executive Order number be identified
for the system to be installed, regardless of the area of
the country.
While most of the permit forms and application
requirements are similar, the procedures vary immensely
after the submission of the application. Due to resource
restraints, each air pollution agency must determine the
focus of their Stage II program. Invariably, programs are
concentrated either on permitting or inspections.
Therefore, the criteria for the issuance of operating
permits can range from a paperwork type exercise, with
emphasis on inspections, to permitting requirements based on
stringent testing.
The New Jersey DEP receives the application; checks to
confirm that all information is complete and that the
facility has designated a certified system for installation,
and mails out a permit. The permit contains standard
conditions that leak and pressure decay/liquid blockage
tests must be performed on the system after installation and
that the facility must maintain verification of the tests.
The existence of this documentation is checked during
facility inspections.9
Massachusetts has developed a two-phase compliance
approach. The first phase involves verification that the
appropriate equipment has been installed. This initial
field inspection is described as a "drive by" screening that
defines a minimum level of inspection required to assure
6-17
-------
that installation has occurred. The second phase is the
more detailed verification that the equipment is operational
and is being maintained.10
The San Diego Air Pollution Control District has
perhaps the most stringent permitting and testing program
observed in the country. The program is based on the
experience and knowledge that most emissions from Stage II
equipment are a result of improperly installed systems. The
following is a description of the permitting and testing
program in San Diego.11
An applicant submits an application for a Stage II
permit that contains a preliminary site plan of all tanks,
dispensers, and underground piping. The application is
reviewed in detail by a member of the engineering staff to
confirm that the planned system is in accordance with GARB
certification and San Diego regulations. If all the
preliminary requirements are met, the District grants
Authority to Construct. This Authority to Construct is
issued subject to several requirements. An example is the
applicant must notify the District within 10 working days
after the Stage II installation that construction has been
completed. Temporary authorization to operate begins only
after receipt by the District of this notice of completion
and an "as built" site plan.
The applicant must also have several tests performed
and provide the District with the results. The District
must be contacted within 10 working days of completion of
construction to establish a mutually agreeable test date.
Normally, the tests are witnessed by a District
representative. If the District is not notified of a test,
then this test may be declared invalid, in which case a
retest is required. The required tests are: (1) a pressure
decay/leak test of vapor control system,* (2) a pressure drop
vs. flow test from each nozzle to its associated underground
tanki (3) a liquid test of all vapor piping to ensure
adequate line slope and liquid drainage? (4) a tank vapor
6-18
-------
space tie test to verify the existence of a tank
interconnect vapor pipe; and (5) a maximum dispensing flow
rate determination for at least one nozzle. Each of these
tests is discussed in the following section.
The temporary authorization to operate remains in
effect, unless canceled, until the facility is inspected by
the District for a Permit to Operate. If the facility
passes inspection, written authorization is given for
continued operation, which is followed by issuance of the
Permit to Operate. The above tests are required to be
repeated if the Stage II piping or equipment is changed in
any way.
6.3.3 Testing Requirements
While efficiency testing is not practical for each
service station, there are tests that indicate improper
installation of underground Stage II vapor piping. These
tests are the pressure decay/leak test, the dynamic back-
pressure test, and the liquid blockage test. Testing
requirements are usually included as a permit condition but
could be specified in the regulation. Various test methods
are contained in Appendix J.
6.3.3.1 Pressure Decav/Leak Test. This test procedure
is used to quantify the vapor tightness of any vapor
recovery system installed at a gasoline dispensing facility.
Leaks in a balance system can cause excessive vapor
emissions. Leaks in an assist system can decrease the
efficiency of the vapor collection or processing system, or
cause the pumps and the incinerator to operate continuously
while attempting to maintain pressure or vacuum.
The test is conducted by capping the vent pipe(s) and
pressurizing the vapor piping system with nitrogen. This
pressurization can be accomplished by introducing nitrogen
into the vapor passage at one nozzle but is commonly done at
the riser in the dispenser. An initial pressure of 10
inches water column is obtained and the final pressure in
the system is recorded after a period of 5 minutes. The
6-19
-------
final pressure is compared to minimum requirements linked to
the ullage space in the tank. Example test procedures of
this type are contained in Appendix J, Sections J.l and J.5.
6.3.3.2 pynamic Pressure Drop Test. This test is used
to determine the pressure drop (flow resistance) through
balance vapor recovery systems (including nozzles, vapor
hose, swivels, dispenser piping, and underground piping) at
prescribed flow rates. The test method consists of flowing
gaseous nitrogen through a calibrated test panel into the
vapor recovery system at different flow rates to simulate
the back pressure created during vehicle refueling. The
resulting backpressures are measured near the nozzle
faceplate using a pressure gauge, and compared with CARS
certification criteria. The system passes this test if, at
the nitrogen flow rates of 20, 60, and 100 SCFH, the flow
resistance measured does not exceed 0.15, 0.45, and 0.95
inches of water, respectively. This test should be run on
every nozzle because nozzles, hoses, and dispenser
connections can cause excessive backpressure. However, in
the event of limited resources to run this number of tests,
the proper approach would be to run this test at a minimum
of the farthest dispenser from the underground tanks for
each product grade. The test procedures in Appendices J.2
and J.4 are for this test.
6.3.3.3 Liquid Blockage T,est. This test is used for
balance and assist systems to determine if the piping
configuration is correct and to detect low points in the
piping where the accumulation of liquid condensate may cause
blockages which restrict the flow of vapors and thus
decrease the system's vapor collection efficiency. The test
method consists of introducing gasoline into the vapor
piping at any point up to and including the riser. When
adequate time has been allowed for the gasoline to flow back
to the underground tank, gaseous nitrogen is introduced into
the vapor piping at the three flow rates of 20, 60, and 100
SCFH. A liquid blockage is indicated either by the needle
6-20
-------
pegging on the pressure gauge and/or wild pulsing of the
needle, or a reading in excess of the limits discussed above
using the dynamic pressure drop test apparatus. This test
is conducted using the same test methods contained in
Appendices J.2 and J.4.
6.3.3.4 Vapor Space Tie Test. An addition to the leak
test/pressure decay procedure discussed above allows the
determination of whether all underground tanks are plumbed
into the system. After the pressure drop has been measured
for the specified time period, the dry break on each
underground tank fillpipe is depressed. If the tank is
properly tied to the vapor system, a release of pressure
will occur. The absence of pressure in the tank indicates
that the tank is not connected to the vapor piping.
6.3.3.5 Maximum Dispensing Flow Rate Determination.
The dispensing flow rate may be checked by simply noting the
volume of gasoline pumped in a specific time interval. This
can be done during the fueling of any vehicle. This test
procedure is contained in Appendix J.3.
6.3.3.6 Liquid Removal Device Test. In addition to
the tests required in San Diego, there is also a mass draft
test method to check liquid removal devices in the hoses.
This test can be performed to check the operation of this
device. It is conducted by introducing sufficient gasoline
into the vapor passage of the coaxial hose to produce a
dynamic back-pressure between 2.0 and 6.0 inches water
column. This is accomplished with approximately 150 ml of
gasoline. Then approximately 10 gallons of gasoline are
dispensed into a vehicle fuel tank. The liquid remaining in
the vapor passage is then drained and the volume is
measured. If the device is operating properly, most of the
gasoline should be removed from the vapor passage during
this fuel dispensing.
6-21
-------
6.4 INSPECTIONS
The emphasis of most Stage II programs is on the
inspection program. The utilization of approved or
certified equipment and the maintenance of this equipment is
essential to the effectiveness of a Stage II vapor recovery
program. Therefore inspection procedures and frequency,
inspector training, and the method of handling violations
are enforcement related matters that need serious
consideration. Unfortunately, most inspection programs
concentrate on the above ground portion of Stage II systems,
with little or no attention given to the underground piping.
Testing procedures can also be incorporated into the
inspection program.
6.4.1 Inspection Checklists and Procedures
Detailed inspection procedures and checklists are
helpful in the development and implementation of a
consistent and equitable enforcement program. All of the
standard agency pre- and post-inspection procedures such as
identification of the purpose of the inspection and
consultation with the owner/operator after the inspection
should be followed. In addition, procedures specific to the
inspection of Stage II equipment can be developed. The
Compliance Assistance Program of GARB publishes a Technical
Manual for Inspectors of Gasoline Vapor Recovery systems.12
The inspection procedures shown in Table 6-2 are taken from
this document, and describe step-by-step instructions for
inspecting Stage II equipment at a gasoline dispensing
facility. Also, Appendix K contains various inspection
checklists and inspection procedures from other areas.
6.4.2 Inspection Frequency
The inspection frequency also varies among different
agencies. The inspection frequency is a direct reflection
of the resources allocated for a Stage II program. The
frequency ranges from one inspection per facility every 5
years to two or three annual inspections per facility.
There is a correlation between inspection frequency and the
6-22
-------
TABLE 6-2. PHASE II INSPECTION PROCEDURES
1. Fueling instructions:
a. See that fueling instructions are clearly displayed with
the appropriate toll free number.
2. Nozzles:
a. Check each nozzle to verify that it is a current GARB
certified model.
b. Verify that each nozzle is installed in accordance with
ARE Executive Orders.
c. Check to see that required nozzle components are in
place and in good condition. Check:
1) required nozzle components (See 401.3.1).
2) automatic shut-off mechanism (observe the filling of
vehicles look for signs of spillage.
3) trigger (is it leaking or broken)
4) spout for damage or looseness (wiggle the spout)
5) leaded nozzle or spout to ensure that it has not been
replaced an unleaded nozzle or spout (check the
diameter).
6) nozzle for leaking gasoline or vapor (tip the nozzle
down into a container and look for vapors).
3. Faceplate:
a. Make sure that the faceplate is smooth, uniform, and
capable of forming a tight seal for balance system and
in good working order for assist systems.
4. Bellows:
a. Stretch the bellows to check for holes, rips, or tears.
b. Check to see that the bellows is securely attached to
the nozzle.
Check to see that the shape of the bellows is normal and
that there are no deformities.
5. Spring:
a. Check to see that the internal bellows spring is not
missing, broken, distorted, welded, or homemade. Many
of the newer balance systems do not require the internal
spring.
6-23
-------
TABLE 6-2. PHASE II INSPECTION PROCEDURES (CONTINUED)
6. Latch:
a. Check to see that the latching device is not missing,
broken, distorted, welded, or homemade.
NOTE
Neither the spring nor the latching device is required on
the Hasstech system, but either may be present. Both the
spring and latching device are required on the Hirt system.
The Amoco bellowless nozzle incorporates a tightly wound
spring around the spout as a latching device.
7. Check valve:
a. See that the check valve is in place (inspect the nozzle
for sign of tampering)
8. Hoses:
a. Only coaxial vapor recovery nozzles and hoses may be
installed on balance systems after February 20, 1986.
Hose configurations must be in compliance with the
exhibits in the most current version of executive order
G-70-52.
b. Check to see that product and vapor hoses with the
overhead retractor are long enough to permit natural
drainage into vapor return piping when the retractor is
in the retracted position, but still avoid kinking when
fully extended.
c. Check to see that hoses with retractors are adjusted to
maintain a proper loop, and that the bottom of the loop
is within the distance from the island surface certified
by the ARE Executive Order for that particular dispenser
configuration.
d. Check to see that hoses are not torn, flattened or
crimped.
e. See that the vapor recovery hoses are of the required
size and length.
f. If liquid removal device is required, check to see that
it is properly installed.
6-24
-------
TABLE 6-2. PHASE II INSPECTION PROCEDURES (CONTINUED)
9. Flow Limiter:
a. If required, open the dispenser (get the key from the
owner or operator) and check to see that the flow
limiter indicator arrow is pointing in the same
direction as the flow of gasoline and that the flow
limiter is not missing.
10. Swivels:
a. Nozzle and dispenser swivels are optional with the
lightweight coaxial hoses for many configurations.
Check the appropriate executive order to see what
swivels are required.
b. Check to see that swivels are lubricated to maintain
power movement (look for full movement).
c. Check to see that swivels are not missing, defective, or
leaking,
d. Check to see that the dispenser end swivels are Fire
Marshal approved. (look for the Fire Marshal sticker).
11. Vent Pipes Pressure Relief Valve
a. Observe to see that the valve is in place if required
for a vacuum assist system.
12. Vacuum Pump (Amoco Bellowless System Only)
a. Wait for a vehicle to fuel.
b. Verify that fuel is being dispensed into the vehicle by
checking the flow meter on the dispenser. Listen toward
the top of the dispenser for a rapid "clicking" sound of
the vapor pump. The "clicking" is caused by the
movement of the pump seals as they rotate within the
pump housings. Clicking sounds indicate that the pump
is working properly.
13. Collection Unit (Hasstech Only):
a. Wait for a vehicle to fuel.
6-25
-------
TABLE 6-2. PHASE II INSPECTION PROCEDURES (CONTINUED)
b. Go to the collection unit and listen for the sound of
the vacuum/blower inside the collection unit. If the
collection unit does not appear to be operating, check
to see that the power switch is ON. If the switch is ON
and the collection unit is still operating, check the
control panel.
14. Control Panel (Hirt system only)
a. Check to see that the power switch is in the on
position.
b. Check to see that both the power and vacuum lamps are
illuminated.
If power lamp is out:
1) Check to see that the on/off switch if on.
2) Check to see that the circuit breakers in the main
electrical panel box are on.
If the vacuum lamp is out:
1) switch the vacuum and power lamp bulbs to verify that
the vacuum lamp is not burned out.
2) check to see that all fill caps and Phase I vapor
recovery connections are on and are tightly sealed.
15. Processing Unit:
a. Look for convection currents coming out of the burner
stack on top of the processing unit, indicating that the
burner is operating (the burner will not be operating at
all times). You may be able to see these currents more
easily by standing back and observing the top of the
stack against a background (such as power lines) or by
looking for the shadows on the ground.
16. Vacuum gauge (Hirt Only):
If the vacuum pump is illuminated, there is no need to check
the vacuum gauge. If the vacuum lamp is not illuminated, a
check of the vacuum gauge is needed.
6-26
-------
TABLE 6-2. PHASE II INSPECTION PROCEDURES (CONTINUED)
The vacuum gauge may be found inside the base of the
dispenser furthest from the vent risers.
a. If the gauge reads zero or negative during dispensing
and non-dispensing, the system is operating okay.
b. If the gauge reads positive during non-dispensing or
pegs to positive during dispensing, the system needs
attention.
Source: CARB Technical Manual for Gasoline Facilities? Phase I
and II, CARB Compliance Assistance Program.
6-27
-------
number of defects found, although there are other relevant
factors.
San Diego inspects private facilities once per year and two
or three times per year for retail service stations.13 The
retail facilities in the Bay Area are inspected twice per year
and the private facilities once per year.14 In the South Coast
District, they strive to average two inspections per year per
facility. However, their inspection program is not geared to
inspect each station twice annually, but rather is a priority
inspection program. Stations which have exhibited recurrent
problems in the past are inspected three times per year, average
situations twice per year, and very conscientious stations are
inspected only once per year. Also, South Coast is experimenting
with a "self inspection" program in which larger companies
implement their own inspection program and report to the
District. Preliminary assessments are encouraging, but an
overall evaluation of this program has not been conducted.15
6.4.3 Inspector Training
The level of training for Stage II inspectors also varies
widely. It is critical that inspectors understand Stage II
technology fully to be able to recognize violations and potential
problems. While segments of the inspection procedures are
relatively simple, such as the identification of torn bellows and
hoses, items such as proper check valve function and the
identification of properly certified equipment cannot be grasped
in a short training program.
Inspector training ranges from agencies that provide a 2-4
hour discussion which includes a video of inspection procedures
to those which have a training program that lasts up to 7 weeks.
The Evaluation and Training Section of CARB has a series of
training courses for inspectors. Generally, inspectors attend a
2-day training course that includes detailed discussion of
equipment technology, CARB certification procedures and Executive
Orders, inspection techniques, test procedures, and a hands-on
section in the field. CARB believes that this 2-day
workshop/training event could easily be 3 or more days to
6-28
-------
adequately cover the necessary material.16 The South Coast
District has a 7-week district training program which includes
working with an experienced inspector for 2 weeks. They also
have training videos on inspection techniques.17
There are currently two videos used most often by State and
local agencies. These are "Stage II Controls", by Multinational
Business Services (MBS) in Washington D.C. and "For Cleaner Air:
Vapor Recovery" by CARB.
6.4.4 Testing Purina Inspection
As mentioned previously, Stage II inspections often focus
entirely on the above ground portion of the system. The
inspection procedures taken from the CARB technical manual that
are cited above include no mention of underground piping testing.
However, the pressure vs. flow and liquid blockage tests can be
conducted by inspectors in the field with minimal time and
effort, and they can provide an idea of the condition of the
underground piping. As discussed in Chapter 4, liquid blockages
can severely inhibit the emission reduction from Stage II systems
even when all nozzles, hoses, and above ground equipment are well
maintained. This testing during inspections is especially
critical for programs that do not require testing during the
permitting process.
The Bay Area District has testing units available for use by
their inspectors. Tests are conducted on a random type basis
during normal inspections and in response to complaints that seem
to indicate liquid blockage type problems.18 Without exception,
every California official with knowledge and experience in Stage
II technology interviewed by EPA indicated that the testing of
the underground piping for leakage and liquid blockage is
possibly the most important aspect of the functioning of Stage II
systems.19
6.4.5 Violations
There are two basic methods used for handling Stage II
violations. These are removing (i.e., tagging out) defective
equipment from service and administrative penalties for
6-29
-------
violations. Following is a summary of the mandated procedure
that must be followed by all agencies in California.20
When a district inspector determines that a component
contains a defect which substantially impairs the effectiveness
of the system in reducing air contaminants, the district marks
the component "Out of Order". The use of the component is then
prohibited until the component has been repaired, replaced, or
adjusted, as necessary, and the district has reinspected the
component or has authorized use of the component pending
reinspection.
Equipment defects which are considered in California to
"substantially impair the effectiveness of the systems in
reducing air contaminants" are:
(a) Absence or disconnection of any component required to
be used in the Executive Order(s) that certified the
system.
(b) A vapor hose which is crimped or flattened such that
the vapor passage is blocked, or the pressure drop
through the vapor hose exceeds by a factor of two or
more the requirements in the system certified in the
Executive Order(s) applicable to the system.
(c) A nozzle boot which is torn in one or more of the
following manners:
1. Triangular-shaped or similar tear 1/2 inch or more to
a side, or hole 1/2 inch or more in length.
2. Slit 1 inch or more in length.
(d) Faceplate or flexible cone which is damaged in the
following manner:
1. For balance nozzles and for nozzles for aspirator and
educator assist type systems, damage shall be such
that the capability to achieve a seal with a fill
pipe interface is affected for 1/4 of the
circumference of the faceplate (accumulated).
2. For nozzles for vacuum assist-type systems, more than
1/4 of the flexible cone missing.
(e) Nozzle shutoff mechanisms which malfunction in any
manner.
6-30
-------
(f) Vapor return lines, including such components as
swivels, antirecirculation valves and underground
piping, which malfunction or are blocked, or restricted
such that pressure drop through the lines exceeds by a
factor of two or more requirements specified in the
Executive Order(s) that certified the system.
(g) Vapor processing unit which is inoperative.
(h) Vacuum producing device which is inoperative.
(i) Pressure/vacuum relief valves, vapor check valves, or
dry beaks which are inoperative.
(j) Any equipment defect which is identified in an
Executive Order certifying a system pursuant to the
Certification Procedures incorporated in Section 94001
of Title 17, California Code of regulations, as
substantially impairing the effectiveness of the system
in reducing air contaminants.
Where a district inspector determines that a component is
not in good working order but does not contain a defect listed
above, the district provides the operator with a notice
specifying the defect. The owner/operator then must correct the
defect within 7 days or be subject to further action.
Each district in California follows this procedure, although
the imposition of administrative penalties, or fines, varies from
district to district. San Diego assesses a fine for all defects
detected, while other districts impose fines if a certain
percentage of defects is found relative to the number of nozzles,
or if a set number of violations is found.21
California officials note that in some situations this tag
out program has tended to be abused by industry. An extreme
example is the station owner that recognizes equipment is
defective but waits until the inspector tags it out of service,
then immediately replaces it with a new component. A suggestion
from California officials is that any inspection program should
be evaluated carefully to avoid creating the situation where the
inspectors are in effect performing the maintenance program for
the service stations. This can be avoided by making the
penalties substantial enough to ensure that the owner will want
6-31
-------
to find these defects instead of waiting for the inspector to
locate them.22
Other areas impose rather severe fines for any violation
noted by the inspector. In New Jersey, no definition of
malfunctioning or defective equipment is given and much is left
to the discretion of the inspector in this regard. Any defect
noted by an inspector is subject to a fine.23
A mixture of these approaches is being implemented by
Massachusetts. The State requires that the facility tag out
their own equipment if it is found to be defective. If an
inspector visits a site and equipment is tagged and not being
used, then no violation occurs. However, the identification of
defective equipment by an inspector that has not been tagged out
and is being used results in a violation and administrative
penalty.24
Massachusetts also has its own list of violations that
allows an inspector to positively write violations due to the
clarity of this list. In order to set some priority between the
different types of violations which could be detected,
Massachusetts separates the kinds of possible violations into
"potentially emitting" and "non-emitting".25 The description of
these violations, with examples, are shown in Table 6-3.
6.5 SUMMARY
In summary, there are many issues to consider in the
implementation of a Stage II program. The information contained
in this chapter, as well as that provided in Appendix E, will
assist an agency in the initial stages in understanding the
various aspects of planning, permitting, and enforcement that
need attention. In addition, the EPA enforcement guidance
document should be consulted for enforcement guidance and
requirements.
6-32
-------
TABLE 6-3 MASSACHUSETTS STAGE II VIOLATIONS
Title of Violation
Example
PRIORITY, OR "EMITTING"
VIOLATIONS
1. Dispensing motor vehicle
fuel without vapor
recovery equipment
2.
3.
4.
5.
6.
Vapor recovery system is
not operating properly
Vapor recovery equipment
is damaged
Failing to prohibit use of
a dispenser with an
inoperative (or
nonexistent) vapor
recovery system
Failing to install signs
to show how to properly
use the vapor recovery
system
Failing to install
certified equipment
7. Failing to perform or mis-
performing a requested
compliance test
Station is not equipped with
Stage II vapor recovery
equipment but is continuing to
dispense fuel.
Bellows has been "tied back",
latch system bypassed,
aspirator not turned on,
processor not turned on.
Could also include a non-spec
configuration (hoses too long
or not assembled correctly)
Tears or holes in the boot,
kinks in the hose, hose is
flattened.
Equipment is damaged but
dispenser is still operational
and could be used.
Signs are supposed to be
conspicuous (outside) and
readable, they must say DO NOT
TOP OFF
Installed equipment is not on
the list of GARB certified
equipment or equipment has
been installed which, although
each piece may be certified,
the components are assembled
in an uncertified
configuration.
Not an immediate concern since
a compliance test would
initially be required only as
a condition of a UAO.
However, if such a request is
made and the facility does not
conduct the test properly, or
ignores the requirement, a
violation would be triggered.
6-33
-------
TABLE 6-3 MASSACHUSETTS STAGE II VIOLATIONS (CONTINUED)
Title of Violation
Example
8. Failing to install and
operate vapor recovery
equipment after the
appropriate deadline
So as to differentiate this
violation from the first
violation type listed above,
the finding of this violation
should be limited to
facilities who have made no
effort to comply with the
requirements of the regulation
(have not filed I&C or R&C
forms) or facilities who are
not listed but still have the
fuel throughput that would
trigger applicability to the
regulation.
OTHER OR "NON-EMITTING"
VIOLATIONS
1. Failing to submit
Installation and
Certification forms
2. Failing to train station
operators
3. Failing to place an "Out
of Order" sign on a
disabled dispenser
4. Failing to maintain
continuous records
Source: Massachusetts Department of Air Quality Control,
Compliance and Enforcement Manual.
6-34
-------
6.6 REFERENCES
1. Stage II Background Information and Technical Support
Document. Massachusetts Department of Environmental
Quality Engineering. January 1989.
2. Telecon. Norwood, P., Pacific Environmental Services,
Inc. (PES) with Wong, R., Dade County Air Pollution
Control. May 14, 1991. Dade County Stage II Program.
3. Memorandum from Norwood, P., Pacific Environmental
Services, Inc. to Shedd, S., U.S. Environmental
Protection Agency. April 30, 1991. Trip Report to
California Agencies to Discuss Stage II.
4. Memorandum from Norton, R. Pacific Environmental
Services, Inc. to Shedd,S., Environmental Protection
Agency. April 29, 1991. Trip Report to New York
Department of Environmental Conservation.
5. Memorandum from Norwood, P., Pacific Environmental
Services, Inc., to Shedd, S., Environmental Protection
Agency. February 22, 1991. Trip Report to New Jersey
Department of Environmental Protection.
6. Telecon. Bowen, E., Pacific Environmental Services,
Inc. (PES) with Wong, P., Dade County Air Pollution
Control. April 15, 1991. Dade County Stage II
Program.
7. Telecon. Bowen, E., Pacific Environmental Services,
Inc. (PES) with Estrusky, B., Pennsylvania DER. April
11, 1991. Philadelphia Stage II Program.
8. Telecon. Bowen, E., Pacific Environmental Services,
Inc. (PES) with Carlson, L., Massachusetts DAQC. April
1, 1991. Massachusetts Stage II Program.
9. Reference 5.
10. Stage II Compliance and Enforcement Manual.
Massachusetts Department of Environmental Quality
Engineering. January 1989.
11. Reference 3.
12. Gasoline Facilities Phase I & II. California Air
Resources Board, Compliance Assistance Program.
Revised March 1991.
13. Reference 3.
14. Reference 3.
6-35
-------
15. Reference 3.
16. Memorandum from Norwood, P., Pacific Environmental
Services, Inc. to Shedd, S.» U.S. Environmental
Protection Agency. April 22, 1991. Trip Report to
Monterey, CA, for CARB stage II Inspection Workshop.
17. Reference 3.
18. Reference 3.
19. Reference 3.
20. California Code of Regulations, Title 17, Section
94006.
21. Reference 3.
22. Reference 3.
23. Reference 5.
24. Reference 10.
25. Reference 10.
6-36
-------
TECHNICAL REPORT DATA
fPleatt n*d tnstructiont on tht revtrte txfort compltnngj
1. RiPORTNO.
EPA-450/3-91-022a
3. RECIPIENT'S ACCESSION NO.
*. TITLE AND SUBTITLE
Technical Guidance - Stage II Vapor Recovery Systems
for Control of Vehicle Refueling Emissions at Gasoline
Dispensing Facilities, Vol. I - Chapters
S. REPORT DATE
November 1991
«, PERFORMING ORGANIZATION CODE
7, AUTHQRISI
8. PERFORMING ORGANIZATION REPORT
9. PERFORMING ORGANIZATION NAME AND AOORESS
US Environmental Protection Agency
Office of Air Quality Planning and Standards
Emission Standards Division (MD-13)
Research Triangle Park, NC 27711
13. PROGRAM ELEMENT NO"'
11 CONTRACT/GRANT NO.
68D10116
12. SPONSORING AGENCY NAME AND AOORESS
US Environmental Protection Agency
Office of Air and Radiation
Washington, DC 20460
13, TYPE OF REPORT AND PERIOD COVERED
Final
1« SPONSORING AQENCV CODE
EPA/200/04
IS. SUPPLEMENTARY NOTES
IB, ABSTRACT
The Clean Air Act Amendments (CAM) of 1990 require the installation of Stage II
vapor recovery systems in ma^ ozone nonattainment areas and direct EPA to issue
guidance as appropriate on the effectiveness of Stage II systems. This document
provides guidance on the effectiveness of Stage II systems and other Stage II tech-
nical information on emissions, controls, costs, and program implementation. Stage
II vapor recovery on vehicle refueling is an effective control technology to reduce
gasoline vapor emissions that contain volatile organic compounds (VOC) and hazardous
air pollutants. Vehicle refueling emissions consist of the gasoline vapors displaced
from the automobile tank by dispensed liquid gasoline. The Stage II system collects
these vapors at the vehicle fillpipe and returns them to the underground storage tank.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTlFIEHS/OPEN ENDED TERMS
COS AT I Field, Croup
Gasoline
Air Pollution
Refueling
Service-Stations
Stage II
Air Pollution Control
«. DISTRIBUTION STATEMENT
Unlimited
19 SECURITY CLASS iTIut Rtparti
Unclassified
NO OP PAGES
212
20. SECURITY CLASS i
Unclassified
22 PRICE
f F* P«r» 2220-1 (••*. 4-77) PMKVIOUS EDITION >• PMSOLKTC
------- |