EPA-AA-SDSB-87-07
Technical Report
Investigation of the Need for In-Use Dispensing Rate
Limits and Fuel Nozzle Geometry Standardization
May 1987
NOTICE
Technical Reports do not necessarily represent final EPA
decisions or positions. They are intended to present
technical analysis of issues using data which are
currently available. The purpose in the release of such
reports is to facilitate the exchange of technical
information and to inform the public of technical
developments which may form the basis for a final EPA
decision, position or regulatory action.
Standards Development and Support Branch
Emission Control Technology Division
Office of Mobile Sources
Office of Air and Radiation
U. S. Environmental Protection Agency
-------
Table of Contents
Section Page
I. Introduction 3
II. Definition of Terms 3
III. Discussion of the Issues 5
A. In-Use Dispensing Rates 6
B. Nozzle Geometries 8
IV. CAA Authority 12
V. Benefits 13
VI. Implementation 14
VII. Economic Analysis 17
VIII. Summary and Conclusions 19
-------
I. Introduction
As a part of the technical information received during the
development of EPA's draft refueling emissions test procedure,
one auto manufacturer suggested that the in-use dispensing rate
of gasoline would have to be controlled and gasoline nozzle
geometries standardized in order to assure the proper in-use
operation of onboard vapor recovery systems (onboard
systems).[!] In response to this suggestion, the effects that
these two parameters (dispensing rate and nozzle geometry) may
have on refueling emissions control were studied. As is
discussed below, this study led to the conclusion that a
dispensing rate limit of 10 gallons per minute (gpm) would be
complementary to the operation of an onboard system and the
control of spillage-related refueling emissions since the
maximum in-use dispensing rate and the dispensing rate for
vehicle refueling emission certification testing would be the
same. However, with regard to the second parameter, additional
information and public comment is needed on the effects of
nozzle geometries on refueling emissions control before it can
be concluded that action is needed.
As a result of the conclusions from this study, the Agency
pursued the development of a voluntary in-use dispensing rate
standard with the American Petroleum Institute (API), which
represents petroleum industry interests. As summarized in
their statement of August 6, 1986, [2] API concluded that it is
not feasible to develop a voluntary standard. API also
concluded that it should be the motor vehicle manufacturers'
responsibility to develop refueling control systems which cause
nozzle shut-off without spitback if design flow limits are
exceeded.
This report presents the results of the investigation of
the effects of dispensing rates and fuel nozzle geometries on
refueling emissions and supports EPA's proposal to regulate
in-use dispensing rates and further study the need for some
form of nozzle standardization. The first portion of the
report is devoted to background. After defining a few key
terms, the report begins with a discussion of the issues. This
is followed by a presentation of EPA's statutory authority to
regulate fuel dispensing rates. After this background
information is presented, the benefits, implementation
approach, and economic impact of a dispensing rate limit are
discussed. The report closes with a brief summary and
conclusions.
II. Definition of Terms
Defined below are some key terms related to fuel
dispensing, fuel dispensing units, and related hardware. Items
related to the fuel nozzle itself are identified in Figure l.
-------
-4-
Fioure 1
Fuel PutQ Nozzle
Nozzle Body
Length of Spout—^
Spout's
Straight —»
Section
Hand Lever
Automatic Shutoff
Port
Automatic Clip
-------
-5-
"Spills" - loss of gasoline during the refueling event due
to one or more of four causes: pre-fill drip, spitback,
overflow, and post-fill drip.[3]
"Spitback" - discharge of gasoline from the fillpipe
resulting from pressure build-up in the vapor space during
an automatic fill.[3]
"Overflow" - loss of fuel from the fillpipe when the
amount of gasoline dispensed exceeds the tank capacity;
often caused by topping off.[3]
"Dispensing unit" - the entire refueling hardware that
exists above ground.
"Pump" - the refueling hardware that exists beneath the
ground or within the dispenser which drives the dispensing
unit.
"Nozzle" - the hand-held portion of the dispensing unit
which consists of the nozzle body, stem, and spout.
"Spout" - the terminal end of the fuel nozzle which fits
into the vehicle's fillpipe when refueling.
"Automatic Shut-off Port" - the hole near the tip of the
spout which activates the nozzle automatic shut-off when
covered with liquid.
With these basic term definitions, it is now possible to
discuss the issues involved in this study.
Ill. Discussion of the Issues
A limited amount of information available in the
literature suggests that both the fuel dispensing rate and
nozzle geometry can affect the amount of emissions generated in
a refueling event. This relationship exists because both the
dispensing rate and nozzle geometry can affect the amount of
turbulent mixing which occurs in the fillpipe, and experience
shows that refueling emissions vary directly with the amount of
turbulent mixing.[4] Thus these parameters may have a small
effect on both the uncontrolled refueling emission rate and the
refueling emissions load to the canister in an onboard vapor
recovery system. Further, in a related area, one source
suggests that gasoline spills during refueling are also
directly affected by the nozzle configuration and dispensing
rate.[3]
The effect of dispensing rates and nozzle geometries on
the control of refueling emissions and gasoline spillage is
discussed below, together with the rationale for any potential
regulatory controls. Dispensing rates are considered first and
nozzle geometries follows.
-------
-6-
A. In-use Dispensing Rates
The fuel dispensing rate is important because of its
effect on the design and performance of an onboard vapor
recovery system. As mentioned above, the fuel dispensing rate
slightly affects the refueling emission rate which is one of
the factors impacting canister size. In addition, since the
fuel dispensing rate determines the rate at which fuel vapor is
displaced from the tank during a refueling event, it also
impacts the design of an onboard system. The major effects are
on the vapor line diameter needed to optimize system
backpressure characteristics, and the size and shape of the
canister needed to capture the refueling emissions.
The dispensing rate also has some effect on the potential
for gasoline spillage (spitback) during a refueling event. If
the fuel is dispensed at a rate greater than vapors can be
displaced from the system, backpressure within the system can
increase and cause a premature shut-off of the fuel nozzle and
the possibility of a fuel spitback.
Since the fuel dispensing rate is one of the key factors
affecting the design of an onboard system, EPA had to establish
a dispensing rate value for use in refueling emissions
testing. Rather than arbitrarily choosing a value, it was
important that the values specified in the certification test
procedure be representative of the dispensing rates expected
in-use.
An investigation of the current in-use dispensing rates by
API revealed that values generally range from 6-12 gpm, with
values normally lying between 8-10 gpm. The in-use dispensing
rates of gasoline varies depending on a number of different
factors. Older and smaller service stations use a suction pump
which is located inside the individual dispenser. These
suction pump dispensers normally operate at 8 to 9 gpm.[5]
Newer, higher volume facilities use submersible turbine pumps
which are located away from the dispensers and are either on or
in the underground storage tank. They serve all the dispensers
and nozzles drawing product from the tank, thus the actual flow
rate varies depending on the number of nozzles being operated
from one pump. These submersible pumps are of a higher
horsepower, and deliver gasoline at rates of 10 to 12 gpm
(presumably when only one or two nozzles are in operation) with
a few newer facilities capable of operating at levels as high
as 15 gpm under the same conditions.[2,5] One in-use survey
reported full serve dispensing rates of 6.5 to 8 gpm with self
service rates slightly higher at 9 to 11 gpm.[5] These rates
were observed at various dispensers and from nozzles of
different manufacturers. All of the available information
indicates that most in-use dispensing rates fall in the range
of 8 to 10 gpm with evidence of a trend toward higher rates in
new stations using higher horsepower pumps.
-------
-7-
Based on this information, the maximum dispensing rate in
EPA's draft refueling emissions test procedure was set at 10
gpm to correspond to values near the higher end of the current
in-use dispensing rate range.[6] The automobiles being
certified will then be designed for and tested at conditions
most comparable to those presented in-use. EPA's draft
refueling test procedure would also essentially reguire that no
fuel spillage occur during the refueling test, since any
spilled fuel which evaporates during the test is included in
the overall emission results. Thus, the implementation of
onboard controls could also result in a reduction in fuel
spillage (spitback) which now occurs when the fuel nozzle
automatic shut-off is activated.
However, if the in-use dispensing rates increase as API
predicts, the dispensing rate used for certification testing of
onboard vehicles will no longer be comparable to those
in-use.[2] Vehicles equipped with onboard systems will not be
refueled in-use at the dispensing rates they were designed for
in certification. Furthermore, many vehicles will not be
designed to accept such high refueling rates and premature
shut-offs and fuel spillage related to spitbacks may increase.
Consequently, the expected reductions in emissions due to
spitback spillage could be decreased or lost totally, and some
of the benefits of controlling refueling emissions may be
offset if spillage increases.
There are basically two options available to prevent an
incompatibility between the certification and in-use dispensing
rates. The first option is to set the dispensing rate used for
certification testing equal to the maximum rate reasonably
expected in-use in the future, rather than 10 gpm. This would
ensure that the in-use dispensing rates do not exceed the rate
used for certification, but it could increase the cost of
onboard systems because they would have to be designed to
handle much higher flow rates. This option also allows in-use
dispensing rates to increase even further in the future since
the vehicles would be able to accept these higher rates. Thus
incompatibility between certification and in-use dispensing
rates could reappear in the future, and the problem would not
be solved.
The second option is to implement an in-use dispensing
rate limit which is equal to the maximum rate used in the
certification testing of onboard-equipped vehicles. As
discussed above, this value would be 10 gpm. Under this
option, retail gasoline marketers and wholesale
purchaser-consumers would be required to limit the dispensing
rate .of their gasolines to 10 gpm. The maximum dispensing rate
chosen for certification testing (10 gpm) is based on values
found near the high end of the current in-use range, so most
service stations would not be affected except to restrict any
-------
-8-
future changes. This option would ensure that the
certification and in-use dispensing rates remain comparable in
the future. This compatibility would enhance the overall
effectiveness of the onboard refueling controls and achieve
additional control of refueling emissions by reducing the
amount of in-use spillage.
Based on the discussion above, the .second option is more
favorable in assuring that the maximum certification and in-use
dispensing rates remain equivalent. Establishing an in-use
dispensing rate limit equal to the maximum certification rate
(10 gpm) would guarantee that the rate used in the design of
the onboard system for certification testing is comparable to
that in-use. This would complement the performance of onboard
controls and reduce in-use spitback spillage. In addition, an
in-use dispensing rate limit will prevent increasing dispensing
rates as vehicles that are designed to handle higher flow rates
enter the fleet, as may have occurred under the first option
considered. As will be discussed further below, this option
could also provide additional environmental benefits by
decreasing the amount of gasoline spillage which occurs during
the refueling of vehicles without onboard controls.
B. Nozzle Geometries
As was mentioned previously in the introduction, at least
one automobile manufacturer has expressed concern to EPA that
without uniform nozzle geometries mechanical fillpipe seals may
not be feasible. They feel such design standards should
include at minimum an in-use dispensing rate limit, as well as
specifications on length, diameter, position and angle of
nozzle bends, position of latching mechanisms and control on
burrs and protrusions that could potentially damage the seal
mechanism and harm the onboard control efficiency.[1] In a
broader context, the SAE Fuel Supply Systems Subcommittee has
also expressed similar concerns regarding standard nozzle spout
specifications. At a recent meeting of this Subcommittee,
several auto manufacturers conveyed a belief that such nozzle
standards are needed to ensure proper interface between the
nozzle spout and auto fillpipe.[7]
In response to these concerns, EPA investigated the
effects of fuel nozzle geometries. During this investigation,
it was learned that nozzle geometries may have an effect on
spillage. First, according to a study by Scott Research
Laboratories, the nozzle bending angle and spout length are two
factors affecting spitback spillage.[3] These can cause
increased splashing against the fillpipe wall and greater
turbulence.[4] This splashing and turbulence could cause more
frequent premature nozzle shut-offs. The greater number of
premature nozzle shut-offs results in an increase in the amount
of spitback spillage which occurs during refueling operations.
-------
-9-
Second, the bending angle of the spout affects the fuel
flow to the vehicle tank within the fillpipe. The increased
splashing against the fillpipe walls and greater turbulence
created by some bending angles prevent optimum fuel flow.[4]
Due to this variance in angles, some fillpipes may not be able
to achieve their design fill rate without premature shut-offs.
Therefore, standardized nozzle geometries could ease the
automobile manufacturers' design of fillpipes which ensure
optimum fuel flow to the vehicle tank and minimize premature
shut-offs.
Third, since nozzle geometries also affect the location of
the automatic shut-off port within the fillpipe, varying nozzle
and fillpipe designs can impact the potential for fuel spitback
when the automatic shut-off device is activated at the end of a
refill. The closer the shut-off port is to the fillpipe
outlet, the higher the fuel must rise in the fillpipe to
activate the shut-off, and thus the greater the chance for a
fuel spitback to occur.
In addition, this investigation studied the currently
available fuel nozzles. As is discussed in more detail below,
current nozzle configurations and geometries were investigated
and found to be relatively similar. This similarity in
configurations is sought primarily for marketing reasons.
Underwriters Laboratories examines the compatibility of
different manufacturers' nozzles with listed power-operated
dispensing units and labels the nozzles as "interchangeable" if
compatible.[8] All "interchangeable" nozzles can be used on
any listed dispensing unit. Therefore, in order to be
"interchangeable" and more marketable, the design of the spout
and handle location of most nozzles are comparable, but not
necessarily identical.
Furthermore, EPA regulation and some voluntary standards
for nozzle designs also contribute to the similarity of current
nozzle geometries. These regulations and standards include the
following:
Federal Register - 40 CFR 80.22
This ruling regulates the outside diameter for both
leaded and unleaded nozzles spouts. It also
specifies a minimum straight section length and
position of the retaining spring for unleaded nozzle
spouts.[9]
UL 842 - "Valves for Flammable Fluids"
This standard of Underwriters Laboratories specifies
the strength and endurance of the nozzle material.
It also gives a maximum nozzle length.[10]
-------
-10-
• NFPA No. 30 - "Flammable and Combustible Liquids"
This standard of the National Fire Protection
Association specifies that any nozzle dispensing
Class I liquids such as gasoline must be automatic
closing and listed by Underwriters Laboratories (and
therefore conform to UL 842).[11]
Compliance with the EPA regulation ruling is mandatory for
all nozzle manufacturers, whereas most manufacturers choose to
adopt the other voluntary standards presented above solely for
insurance/liability purposes. It is important to emphasize,
however, that to date no real limits on the flow rates or
configuration, besides diameters and a straight section length,
have been made. The manufacturers have total freedom in their
nozzle designs, except for market and insurance constraints.
The International Standards Organization (ISO) is
currently evaluating possible design standards for all fuel
nozzles in order to reduce fire risks. These draft standards
include specifications on the dimensions and geometries for
nozzle spouts as well as a maximum dispensing rate.[12]
Response to these proposed ISO standards by SAE and other
American concerns has not been positive. They believe that
these standards specify too many parameters and are therefore
too restrictive. Consequently, the U.S. voted to disapprove
the proposed ISO standards.[13]
As part of this investigation of fuel nozzle geometries,
some key dimensions of nozzle spouts from the two largest
nozzle manufacturers were compared.[14,15] The designs of
these two nozzles are followed closely by other manufacturers
and these nozzles are often rebuilt. Thus they are quite
representative of the majority of in-use nozzles. These fuel
nozzle design characteristics of the two manufacturers are
presented in Table 1. This table shows that in general the two
nozzle spouts are similar with slight differences in the
straight section length and dispensing rate characteristics.
Since the nozzle configurations are quite similar, nozzle
standardization may not be necessary. It is not clear that
minor differences such as these would impact the operation of
an onboard system, although there is some evidence that nozzle
geometries could very slightly affect the refueling emission
rate due to varying amounts of turbulent mixing and droplet
entrainment.
However, even if nozzle geometries are standardized, it is
not clear than the desired effect would be achieved. The
nozzle spout position within the fillpipe is also affected by
the angle at which the individual refueler inserts the nozzle
itself during each refueling event. The variability in nozzle
position caused by this effect alone could be enough to negate
any potential emission or control system operation benefits
which arise from nozzle standardization.
-------
-11-
Table l
Fuel Nozzle Design
Dimension
Total- Spout Length
Spout Straight Sec . Length
Spout Bending Angle
Automatic Shut-off Location*
Nozzle Outside Diameter
Dispensing Rate @ 5 psi AP
@ 10 psi AP
Characteristics
OPW 11 -A EMCO
7"
3.9"
24-25°
0.73"
.81"
4gpm
lOgpm
Wheaton A2000
7.25"
2.9"
27°
0.67"
.81"
7 . 5gpm
13 . 5gpm
* Distance from nozzle tip.
** Maximum.
-------
-12-
Given this effect and the relative uncertainty regarding
the impact of nozzle geometries on refueling emissions and
their control, it is not clear that nozzle geometries need to
be standardized to control refueling emissions. If further
information becomes available which demonstrates that the
impacts of varying nozzle geometries on refueling emissions and
effects on onboard control designs are significant and that the
costs of nozzle standardization are reasonable, then such
standardization will be considered.
Based on the rationale presented above, it appears that
in-use dispensing rate limits would enhance the effectiveness
of onboard controls and reduce spitback spillage. Standardized
nozzle designs may be considered in the future, but further
study and information is required. The remainder of this
report provides further background information and analysis
pertaining to a dispensing rate limit requirement.
IV. CAA Authority
The Clean Air Act (CAA) gives EPA authority to regulate
the in-use dispensing rate. EPA's authority in this area stems
from Section 211 (c) of the Act. This section allows EPA to
control or prohibit the introduction into commerce or offering
for sale of any fuel for use in a motor vehicle or motor
vehicle engine if the emission products of the fuel contribute
to air pollution which endangers public health or welfare or
the emission products of such fuel will impair to a significant
degree the performance of any emission control system. In this
case, the emission products of the fuel are the refueling
vapors and evaporated spilled fuel associated with the
dispensing of gasoline. These vapors are photochemically
reactive hydrocarbons which contribute to the formation of
ozone, an NAAQS criteria pollutant. The emission control
system of interest in this case is an onboard vapor recovery
system. As discussed previously, dispensing rate limits could
enhance the efficient performance of an onboard system and
avoid significant impairment in the control of refueling
emissions by onboard controls through a reduction in spillage.
Section 211 (c)(2)(B) further requires EPA to consider all
available scientific and economic data on possible
alternatives, including a cost benefit analysis comparing
emission control systems, when regulating a fuel in order to
protect the effectiveness of emission controls. The
alternative to onboard refueling controls is Stage II vapor
recovery systems. This alternative has been thoroughly
analyzed by EPA. Although Stage II is a technically feasible
alternative, EPA believes that vehicle-based vapor recovery, or
onboard systems, is the preferred control approach.
Regardless, Stage II systems would most likely also require
-------
-13-
flow rate limits to ensure proper performance as is now
required in California.[16] Thus, a limit on in-use dispensing
rates appear consistent with the requirements of Section 211(c)
of the CAA.
V. Benefits
A number of benefits can be achieved by limiting in-use
dispensing rates. These include enhanced control of refueling
emissions, gasoline savings, plus health and safety benefits.
Each of these is discussed below.
First, as previously mentioned, a dispensing rate
requirement would complement the efficient operation of an
onboard refueling control system. With a dispensing rate
limit, the dispensing rate used for the design of an onboard
system would be compatible with the rates observed in-use.
Second, a dispensing rate limit would provide both
refueling emission reductions and gasoline savings by reducing
in-use spillage. A dispensing rate limit would reduce or
practically eliminate premature nozzle shut-offs and spitback
from onboard-equipped vehicles and reduce the spitback from
current in-use vehicles without onboard controls. The onboard
vehicles would be designed to accommodate 10 gpm dispensing
rates without spitback. For in-use vehicles without onboard
controls, a dispensing rate limit would reduce spillage for
those vehicles which can handle a flow rate of 10 gpm, but not
higher values which could occur if in-use dispensing rates were
allowed to increase. However, some in-use vehicles cannot
handle even a 10 gpm dispensing rate; a dispensing rate limit
would not have any effect on the spitback from these vehicles.
The current average emission factor associated with
spitback spillage is 0.15 g/gallon of dispensed fuel.[3] Using
this emission factor and EPA's fuel consumption projection for
1995, spitback spillage is estimated to cause about 11,000 tons
of refueling emissions in that year.[17] This spitback
spillage amounts to 3.7 million gallons of gasoline spilled in
1995. In 1995, the vehicles without onboard controls will
comprise about 50 percent of the total vehicle fleet, if
onboard controls are implemented for 1990 model year
vehicles.[18] Assuming that a dispensing rate limit would
eliminate spitback from all vehicles with onboard systems and
one-half of non-onboard equipped vehicles, and a gasoline price
of $1 per gallon, such a dispensing rate limit could provide a
savings of approximately $2.8 million for gasoline consumers in
1995. This provides an emission reduction of 8,250 tons per
year. In addition to these emission reductions and gasoline
savings, a dispensing rate limit would provide additional
convenience to the general public by reducing spillage of
gasoline on automobiles or persons.
-------
-14-
It is important to note that the current spitback spillage
emission factor of 0.15 g/gallon of dispensed fuel is probably
conservative. This emission factor originated from a 1972
study done by Scott Research Laboratories. Since most stations
in 1972 used suction pump dispensers, which have lower
dispensing rates than the current stations using submersible
turbine pumps, it would be reasonable to assume that the in-use
dispensing rates at that time were lower than present
rates.[2] Since the study found a higher probability for
spitback spillage at faster dispensing rates, today's spillage
emission factor may be greater. Furthermore, at the time of
the study, most stations were full service and only gasoline
attendants, people with greater experience dispensing gasoline
than today's self-service customers, were used in this
analysis. Hence, it is reasonable to believe that today's
self-serve customers may spill gasoline more frequently and in
larger amounts than the gasoline attendants used in the 1972
study. Another important factor in this Scott study is the
presence of a person monitoring the gasoline attendants. With
a monitor present, the attendants could have been more careful
and less likely to spill gasoline. For these reasons, the 0.15
g/gallon emission factor used in this report for spitback
spillage may be conservative.
Third, establishing an in-use dispensing rate limit will
provide some health benefits for the general public. A
dispensing rate limit will help reduce the ozone concentration
in the lower atmosphere by enhancing control of refueling
emissions. In addition, repeated or prolonged dermal contact
with liquid gasoline due to spillage can cause irritation and
dermatitis for some individuals.[19] Thus, there are some
health benefits gained from limiting the in-use dispensing
rates.
In addition to the health benefits, a dispensing rate
limit provides some additional safety benefits for refueling
operations. The refueling operation would be inherently more
safe because of reduction in spitback and spillage. An in-use
dispensing rate limit would help to reduce spillage during
refueling, and would thus contribute to a reduction in the
potential for fires caused by inadvertent ignition of spilled
gasoline.
VI. Implementation
Any regulation to control dispensing rates should be
implemented with minimal impact on service station operations.
With this in mind, the possible approaches for limiting in-use
dispensing rates were evaluated. The implementation approaches
are discussed below including the necessary liability and
enforcement aspects. However, one of the key factors affecting
how dispensing rate limits could be regulated is the nature and
structure of the nozzle industry. This is discussed first.
-------
-15-
Since an in-use dispensing rate limit and/or nozzle
geometry standards would have some impact on the nozzle
industry, it is important to highlight its structure. The fuel
nozzle industry consists of both primary manufacturers and
rebuilders. The primary nozzle manufacturers produce equipment
from all new materials whereas rebuilders reuse the original
nozzle body castings and replace all the old and worn operating
parts. According to the Petroleum Equipment Institute,
approximately 70 percent of the entire market consists of
originally manufactured nozzles.[20] Moreover, the nozzle
market structure consists primarily of a few large original
manufacturers, a few mid-size original manufacturers who also
rebuild nozzles, and several small rebuilders.
The rebuilders, or remanufacturers, range greatly in
size.[20] Some rebuilders operate from their trucks and travel
to service stations replacing worn-out nozzle parts. Other
rebuilders are larger companies who rebuild and test the
nozzles they rebuild to meet given specifications. Recently,
primary manufacturers have become more concerned about the
liability of rebuilt nozzles since they could be held
responsible for a failure of a rebuilt nozzle with their
original casting.[20] Because of this concern, one original
manufacturer has quit rebuilding nozzles and has tried to
prevent others from rebuilding their original nozzles.
With respect to the implementation of an in-use dispensing
rate limit, it is important to characterize what causes the
variation in dispensing rates. The flow rate from the nozzle
spout is dependent on two factors: the pressure supplied to
the nozzle by the pump and the pressure at the orifice created
by the hand lever position on the nozzle. These combine to
form a pressure drop which governs the flow rate through the
nozzle. The same nozzle can produce a variety of flow rates
depending on this pressure drop. For any given nozzle
configuration, the maximum pressure drop is controlled by the
supplied pump pressure since the nozzle characteristics are
fixed. The maximum dispensing rates of different
manufacturers' nozzles vary with the supplied pump pressure due
to different nozzle characteristics. Thus, to control the
maximum flow rate through the nozzle, the resulting pressure
drop created by the pump and the lever position must be
controlled. This could be done by controlling the pressure
contribution of both the pump and the lever position. A better
solution, however, is using a variable flow limiting orifice or
flow restrictor to limit the flow rate. This would control the
wide variance in dispensing rates that could occur with
different combinations of pump pressures and nozzle
configurations.
-------
-16-
As mentioned above, in order to limit the dispensing rate
a flow restrictor or limiting orifice would have to be
incorporated somewhere in the dispensing system. This flow
restrictor could be located in the dispensing unit, the hose,
or the nozzle itself. The in-use effectiveness of the
dispensing limit regulation may depend to some degree on the
restrictor location. The best location would be one where the
restrictor provides effective dispensing rate control, easy
installation and resistance to tampering. Inserting the
restrictor or limiting orifice internal to the nozzle body
appears to be the best way to achieve these requirements and
thus may be the best approach to control dispensing rates.
There are basically two implementation approaches for an
in-use dispensing rate limit requirement. First, all gasoline
retailers and wholesale purchaser-consumers could be required
to limit dispensing rates at the nozzle to no greater than 10
gpm. It places responsibility for compliance with the
dispensing rate limit on the gasoline retailers and wholesale
purchaser-consumers, those who own and maintain the in-use fuel
nozzles. This option allows that no action be taken if the
dispensing rate limit requirement is being met without any
modifications and also provides flexibility on how compliance
is achieved when measures are needed (i.e., nozzle, pump, or
hose restrictor).
The second implementation method considered would be to
require that nozzle manufacturers design their nozzles so that
the maximum dispensing rate would not exceed 10 gpm. Under
this approach, nozzle manufacturers would probably have to
verify compliance with the dispensing rate regulation through
an EPA certification program. The gasoline retailers and
wholesale purchaser-consumers could then purchase certified
nozzles for dispensing gasoline at their facilities. This
alternative would guarantee that the flow regulator was part of
the internal nozzle design and also would insure the
availability of a product which conforms to the 10 gpm
dispensing rate limit. However, this implementation approach
has several drawbacks. First, it puts at least part of the
liability for controlling in-use dispensing rates on the nozzle
manufacturers, who are not responsible for the maintenance or
condition of the nozzles in use. Second, this option would
force EPA to regulate a new industry, the nozzle manufacturers,
which could have some economic implications on small business.
Third, under this approach it would be necessary to provide a
phase-in period for the effective date of a dispensing rate
limit to prevent a massive turnover of fuel nozzles at the
service stations. Finally, this option would make rebuilders,
which are for the most part small businesses, use a different
approach in order to stay in the market. They would have to
certify each type of nozzle they rebuild. This could
substantially alter their business and marketing operations.
-------
-17-
Based on the discussion above, the most straightforward
approach is the first option which requires that gasoline
retailers and wholesale purchaser-consumers dispense gasoline
at a maximum flow rate no greater than 10 gpm. This approach
has several advantages. First, this option minimizes
compliance costs since most service stations would already be
in compliance and would not have to take any action. If the
maximum flow rate for a nozzle exceeds 10 gpm, however, the
restrictor or limiting orifice discussed above would need to be
placed in the system to limit flow.
Second, this option will not force EPA to regulate a new
industry or have any negative impacts on small business. For
marketing reasons, however, both original nozzle manufacturers
and rebuilders could choose to modify their product designs to
fulfill this in-use dispensing rate limit requirement.
Original manufacturers could either modify the poppet in their
current nozzles to act as a limiting orifice or insert a flow
regulator (variable orifice) in the nozzle. Rebuilders, which
use the original nozzle bodies, could add a fixed or variable
flow regulator if the original nozzle was not already designed
to comply with the dispensing rate limit.
Third, this option is preferable from an enforcement
perspective. Existing enforcement programs at both the state
and federal levels could easily be expanded to include a
maximum dispensing rate measurement. The test procedure used
to determine the flow rate would consist of either an in-line
flow meter or a volume/time measurement.
Finally, the liability for compliance with the 10 gpm
maximum dispensing rate requirement would lie with the gasoline
retailers and wholesale purchaser-consumers. The liability
provisions could be similar to those applied to the current
nozzle spout diameter regulations for leaded and unleaded fuels
(40 CFR 80.23).
VII. Economic Analysis
The economic impact of an in-use dispensing rate limit
requirement is expected to be minimal. Since the 10 gpm value
was chosen from values near the high end of the current
dispensing rate range, most stations are now in compliance
without any additional modifications and will incur no costs.
Therefore, the primary effect of an in-use dispensing rate
limit would be on current service stations which are dispensing
at rates higher than 10 gpm, new service stations, and service
stations which will replace their current underground pumps and
dispensing hardware in the future.
-------
-18-
For those current service stations/nozzles which are
dispensing at rates higher than 10 gpm, some modifications
would be necessary to attain compliance. They could retrofit
their dispensing hardware to satisfy the dispensing rate limit
by inserting a flow regulator in the hose, dispenser, or
nozzle. In the future they could purchase new or rebuilt fuel
nozzles which have a fixed or variable flow regulator and can
effectively regulate dispensing rates.
For new service stations and those stations which will
replace their current underground pumps and dispensing hardware
in the future, equipment they install will have to comply to
the 10 gpm dispensing rate requirement. If their equipment
exceeds the 10 gpm requirement, they could either insert flow
regulators in their equipment (i.e., dispenser or hose) or
purchase nozzles which are designed specifically to control the
maximum dispensing rate to 10 gpm. Furthermore, these stations
would not need to install underground pumps of higher
horsepower than those currently used since the dispensing rate
would be regulated and the increased horsepower would provide
little or no additional benefits. Thus both capital and
operating costs related to the pump could be saved.
Based on the discussion above, it appears that the
simplest and most effective way to regulate the dispensing
rates for for current and future stations would be through
nozzle modifications. The costs associated with these nozzle
modifications are discussed below.
As was mentioned previously, there are two ways to
effectively regulate fuel flow in the nozzle: 1) modify the
existing poppet in most current nozzles to serve as a limiting
orifice, or 2) insert a flow regulator in the nozzle.
Modifying the existing nozzle design would require only minor
tooling changes in the production of the nozzles. These
tooling changes would cost approximately $1 per nozzle to
amortize the necessary fixed costs needed for these changes.
These tooling change costs are only short-term and would be
eliminated after a few years. Inserting a flow regulator,
however, would provide better control of dispensing rates over
a wider pressure range. Based on discussions with nozzle
manufacturers, the flow regulator and necessary tooling changes
are estimated to cost between $1 and $5 per nozzle. [21,22]
Therefore, the hardware and tooling costs for original
manufacturers to incorporate an in-use dispensing rate limit
into their nozzle designs will range from $1 to $5 per nozzle
depending on the type of control the manufacturer chooses to
implement.
Rebuilders have only one alternative to effectively
regulate dispensing rates in the nozzle. This alternative is
to insert a fixed or variable flow regulator in the internal
design of the nozzle. As stated above, the hardware costs
necessary to incorporate this flow regulator into the nozzle
design ranges between $1 and $5 per nozzle.
-------
-19-
Based on this discussion, the economic burden of a
dispensing rate limit requirement would be expected to be very
low. Since the flow rate limit was selected from near the high
end of the current in-use dispensing rate range, most stations
would already meet this requirement and therefore incur no
associated cost. In addition, it is worth noting that any
additional costs to the consumer related to limiting dispensing
rates could be substantially offset by the reduction in
spitback spillage which may result.
VIII. Summary and Conclusions
From the discussion of the issues studied in this
investigation, an in-use dispensing rate limit would enhance
the onboard effectiveness and reduce spitback caused by
premature nozzle shut-offs. Such a limit would ensure that the
maximum in-use dispensing rates are equivalent to the
dispensing rates used in the design and certification of the
vehicle's refueling control system. It may also be beneficial
to specify nozzle geometries to assist in controlling refueling
emissions and prevent spillage. Further information regarding
the impacts of nozzle geometries on refueling emissions and
onboard system designs, however, is needed before such
standardization is considered.
Section 211 (c) of the Clean Air Act gives EPA the
authority to regulate the dispensing rates and geometries of
fuel nozzles. A dispensing rate limit requirement would
enhance the performance and efficiency of onboard controls as
well as provide additional refueling emission reductions and
gasoline savings by reducing spillage. In addition, the
reduction in gasoline spillage will provide several health and
safety improvements for refueling operations. A recommended
approach for a dispensing rate limit would require that all
commercial gasoline retailers and wholesale purchaser-consumers
use nozzles with a maximum dispensing rate of 10 gpm. Federal
and state surveillance teams would have to expand their
enforcement programs to measure the maximum dispensing rate at
service stations. Furthermore, such regulations should have
very small economic impact since most stations would already be
in compliance with the dispensing rate limit.
In conclusion, it appears that development of an in-use
dispensing rate limit is in order to ensure proper functioning
of onboard control systems and reduce in-use gasoline
spillage. The benefits received from such a regulation
outweigh any costs which may result. In addition, further
investigation of the effects of nozzle geometry standardization
should be further studied for the same reasons.
-------
-20-
References
1. Memorandum to R. Fred Holbert from Ford, W. M.
Kreucher, June 17, 1986.
2. "API Response to EPA Request for Comments on
Petroleum Marketing Industry Refueling Rate Voluntary
Standard," August 6, 1986.
3. "Investigation of Passenger Car Refueling Losses,"
Malcolm Smith, Scott Research Laboratories, Inc., September 1,
1972.
4. "A Study of Variables that Effect the Amount of
Vapor Emitted During the Refueling of Automobiles," Edward M.
Liston, Stanford Research Institute, May 16, 1975.
5. Memorandum to Charles L. Gray from American
Petroleum Institute, E. P. Crockett, August 8, 1984.
6. "Summary and Analysis of Comments on the Recommended
Practice for the Measurement of Refueling Emissions," U.S. EPA,
OAR, QMS, ECTD, [MO] 1986.
7. Meeting of SAE Fuel Supply Systems Subcommittee,
Cobo Hall, Detroit, MI., February 23, 1987.
8. "Underwriters Laboratories Gas and Oil Equipment
Directory," Underwriters Laboratories, September 1985.
9. "Controls Applicable to Gasoline Retailers and
Wholesale Purchaser-Consumers," Code of Federal Regulations,
Volume 40, Part 80, Section 22.
10. "Valves for Flammable Fluids - UL 842," Underwriters
Laboratories, June 30, 1980.
11. "Flammable and Combustible Liquids Code," National
Fire Protection Association Code No. 30.
12. "Working Draft: Road Vehicles- Nozzle Spouts for
Unleaded Gasoline," ISO/TC22/SC16/WG3/N24.
13. "Annex with U.S. Disapproval to N74 and N75," R.
Thomas Northrup, Society of Automotive Engineers, November 14,
1985.
14. Product Catalogs and Subsequent Telephone
Conversations with Dover Corp./OPW Division regarding nozzle
spout dimensions, July 1986.
-------
-21-
15. Product Catalogs and Subsequent Telephone
Conversations with EMCO Wheaton regarding nozzle spout
dimensions, July 1986.
16. Performance Test Procedures for Gasoline Vapor
Recovery Systems, Stationary Source Test Methods, February 1984.
17. "MOBILES Fuel Consumption Model," Mark Wolcott,
EPA-AA-TEB-85-2, February 1985.
18. "Evaluation of Air Pollution Regulatory Strategies
for Gasoline Marketing Industry - Response to Public Comments,"
Draft RIA (Vol. 1), November 5, 1986.
19. Patty's Industrial Hygiene and Toxicology, G.D
Clayton and F.E. Clayton, 1978-1982.
20. Telephone Conversation with Howard Upton of
Petroleum Equipment Institute, July 1986.
21. Memorandum to Kathleen Steilen from Husky
Corporation, Arthur C. Fink, Jr., March 3, 1987 and Subsequent
Telephone Conversation on March 6, 1987.
22. Conversation with Alex Podgers of EMCO Wheaton,
Meeting with Nozzle Manufacturers from Petroleum Equipment
Institute, September 23, 1986.
------- |