[Federal Register Volume 61, Number 120 (Thursday, June 20, 1996)]
[Rules and Regulations]
[Pages 31449-31459]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 96-15564]



=======================================================================
-----------------------------------------------------------------------

DEPARTMENT OF TRANSPORTATION

Research and Special Programs Administration

49 CFR Part 192

[Docket No. PS-118; Amendment 192-79]
RIN 2137-AB97


Excess Flow Valve--Performance Standards

AGENCY: Research and Special Programs Administration, (RSPA), DOT.

ACTION: Final rule.

-----------------------------------------------------------------------

SUMMARY: In the process of routine excavation activities, excavators 
often sever gas service lines causing loss of life, injury, or property 
damage by fire or explosion. Excess flow valves (EFVs) restrict the 
flow of gas by closing automatically when a line is severed, thus 
mitigating the consequences of service line failures. In this final 
rule, RSPA has developed standards for the performance of EFVs used to 
protect single-residence service lines. If an EFV is installed on such 
a line, it must meet these performance standards.

DATES: This final rule takes effect July 22, 1996.

FOR FURTHER INFORMATION CONTACT: Mike Israni (202) 366-4571, regarding 
the subject matter of this final rule, or the Dockets Unit, (202) 366-
4453, regarding copies of this final rule or other material in the 
docket that is referenced in this rule.

SUPPLEMENTARY INFORMATION:

Statutory Mandate

    In 49 U.S.C. 60110 Congress directs the Department of 
Transportation to issue regulations prescribing the circumstances under 
which operators of natural gas distribution systems must install EFVs. 
If the Department determines that there are no circumstances under 
which EFVs should be installed, the Department is to report this 
determination, and the reasons for the decision, to Congress. RSPA, on 
behalf of the Department, has determined that there are no 
circumstances under which the Department should require the 
installation of EFVs, primarily because the costs far exceed the 
benefits of such installation. RSPA has sent the report of its reasons 
for this determination to Congress. The report to Congress (April 4, 
1995) and the cost/benefit analysis of mandatory EFV installation are 
available in the docket. Costs and benefits are also discussed later in 
this document under ``Cost/Benefit Analysis.''
    49 U.S.C. 60110 further requires the Department to develop 
standards for the performance of EFVs used to protect service lines in 
a natural gas distribution system. The development of these standards 
is the subject of this rulemaking.
    The statute also requires the Department to issue a rule requiring 
operators to notify customers about EFV availability and to offer to 
install EFVs that meet the performance standards, if the customer pays 
for the installation. RSPA will initiate a separate notice of proposed 
rulemaking for customer notification.

The Problem

    Despite efforts, such as damage prevention programs, to reduce the 
frequency of excavation-related service line incidents on natural gas 
distribution service lines, such incidents persist and continue to 
result in death, injury, fire, or explosion. During the period from 
March 1991 through February 1994, 30 incidents with consequences that 
might have been mitigated by an EFV were reported to RSPA. These 
incidents, mostly excavation-related, resulted in 2 fatalities, 16 
injuries, and an estimated $3,249,595 in property damage. Incident 
history is explained in the November 1991 and January 1995 cost/benefit 
studies evaluating mandatory EFV installation. Because damage 
prevention measures are not foolproof, RSPA has sought to identify ways 
to mitigate the consequences of these incidents. The National 
Transportation Safety Board (NTSB) and others have proposed EFVs as a 
means of mitigation.

NTSB Recommendations

    NTSB has recommended EFVs as a means of reducing or preventing 
injury or death from incidents resulting from service line breaks or 
ruptures. Since 1971, NTSB has issued seven recommendations regarding 
the use of EFVs in service lines. NTSB's recommendations are summarized 
and discussed in the Notice of Proposed Rulemaking on this rulemaking 
(58 FR 21524; April 21, 1993).

The Advance Notice of Proposed Rulemaking (ANPRM)

    RSPA issued an ANPRM (55 FR 52188; December 20, 1990) seeking 
information on the desirability of requiring the installation of EFVs 
on gas distribution service lines to reduce the damage from service 
line ruptures. The ANPRM also contained a questionnaire to collect 
current operational data on the use of EFVs by natural gas distribution 
operators. The results of the

[[Page 31450]]

ANPRM were summarized in the NPRM and are available in the docket.

The Notice of Proposed Rulemaking (NPRM)

    In 1993, RSPA published an NPRM (Notice 2: 58 FR 21524; April 21, 
1993), titled ``Excess Flow Valve Installation on Service Lines,'' that 
proposed to amend 49 CFR Part 192 to require installation of EFVs on 
new and replaced single residence service lines operating at a pressure 
of 10 psig or more. This NPRM also proposed performance standards for 
EFVs and conditions under which EFVs must be installed. The initial 
comment period for this NPRM closed June 21, 1993. The NPRM is 
available in the docket.
    RSPA received 140 written comments in response to the NPRM: 14 from 
industry associations, 1 from an EFV manufacturer, 102 from local 
distribution companies, 2 from consultants, 17 from Congress, state 
agencies, and regulatory associations, 3 from transmission companies, 
and 1 from a group of commenters, designated hereafter as the Joint 
Commenters (see below).

The Public Meeting

    RSPA held a public meeting on June 18, 1993 (58 FR 33064; June 15, 
1993) to enable interested parties to present additional comments on 
several of the issues presented in the NPRM. In the notice announcing 
the public meeting, RSPA also extended the comment period to July 6, 
1993, to allow those not able to attend the meeting to have access to 
the transcript. Representatives of the American Gas Association (AGA), 
UMAC (an EFV manufacturer), the Gas Safety Action Council (GASAC), the 
National Association of Pipeline Safety Representatives (NAPSR), and 
NTSB spoke at the meeting. The AGA representative objected to the 
proposed rule, especially to the expected benefits estimated in the 
cost/benefit study. GASAC, NTSB, and UMAC supported an EFV rule, but 
not as proposed. The NAPSR representative noted that in NAPSR's 
experience EFVs have not been cost beneficial.

The Joint Commenters

    On December 20, 1993, a group, designating itself as the Joint 
Commenters, filed comments that recommended language to include in an 
EFV rule. The Joint Commenters included GASAC, EFV manufacturers, and 
two gas pipeline distribution associations. Although not a signatory to 
the comments, NTSB sent two letters to a pipeline association 
supporting the Joint Commenters' recommendations. The NTSB letters are 
available in the docket.
    The Joint Commenters did not include representatives from the two 
major state pipeline safety associations, NAPSR, and the National 
Association of Regulatory Utility Commissioners (NARUC). NAPSR 
originally participated in discussions with the Joint Commenters but 
later dropped out because NAPSR members oppose a federal requirement to 
install EFVs. The comments from NAPSR are available in the docket.
    The Joint Commenters recommended regulatory language that their 
signatories would support if RSPA were to adopt this recommendation as 
a final rule. In a Notice of Reopening Comment Period, RSPA reopened 
the comment period to solicit comment on the safety merits of the Joint 
Commenters' recommended language (59 FR 39319; August 2, 1994). The 
reopened comment period closed October 3, 1994. In addition to seeking 
comments on the safety merits of the recommendation, RSPA also sought 
comment on: whether to allow EFVs with a bypass feature; whether, and 
to what extent, the presence of contaminants in the gas stream should 
preclude installation of an EFV; and whether RSPA should delay issuing 
a rule until industry performance standards for EFVs are developed.
    An additional 70 comments were received in response to the Notice 
of Reopening Comment Period: 7 from industry associations, 1 from an 
EFV manufacturer, 56 from local distribution companies, 5 from 
Congress, state agencies, and regulatory associations, and 1 from a 
transmission company. A discussion of the 140 comments to the NPRM and 
70 comments to the Notice of Reopening Comment Period and RSPA 
disposition of these comments is found below.

Advisory Committee Review

    The Technical Pipeline Safety Standards Committee (TPSSC) was 
established by statute to evaluate the technical feasibility, 
reasonableness, and practicability of proposed regulations. The TPSSC 
met on August 3, 1993, in Washington, DC, to consider the EFV standards 
proposed in the April 1993 NPRM. The TPSSC voted 11 to 0 against 
adopting the proposed rule as written. In addition, the TPSSC voted 10 
to 1 against RSPA issuing any rule on EFVs. However, the TPSSC voted 10 
to 1 to respect the wishes of Congress and to provide support for the 
Congressional mandate as implemented by RSPA. RSPA addresses each of 
the TPSSC's recommendations in the discussion of comments below.

Petition for Rulemaking

    On July 14, 1995, AGA submitted a petition for rulemaking on EFV 
performance standards and customer notification requirements. In this 
petition, AGA urged OPS to adopt industry performance and manufacturing 
standards as soon as they are available and, in the interim, to adopt 
the performance standards recommended by the Joint Commenters. RSPA is 
not required to consider those comments in the petition pertaining to 
performance standards since the comments were received well after the 
close of the re-opened comment period. However, RSPA notes that those 
comments do not raise any issues not already raised in prior comments 
and addressed in this rule.
    RSPA will consider the bulk of AGA's petition dealing with customer 
notification requirements in the customer notification rulemaking.

Cost/Benefit Analysis (Mandating EFV installation)

    RSPA recognizes the beneficial safety effects of EFVs. However, 
after extensive study and rulemaking, RSPA has decided not to require 
the installation of EFVs, primarily because the costs far exceed the 
benefits of such installation.
    Many comments to the NPRM and Notice of Reopening Comment Period 
cited the need for RSPA to redo the cost/benefit study that had been 
prepared to accompany the NPRM. Commenters said incident frequency, 
fire and police response costs, and property damage costs were 
overstated. The most frequent objection was that RSPA overestimated 
property loss and fire fighting costs for incidents with less than 
$5,000 in property damage. Commenters pointed out that leaks occur with 
greater frequency than incidents and that, by equating leak repair 
reports with incident reports, RSPA overstated the benefits to be 
gained. Many commenters also said that the $20 estimated cost to 
install an EFV was too low.
    In light of the commenters' criticisms, RSPA thoroughly reexamined 
the cost/benefit study. The revised study included updated data 
regarding service line incidents and revised information on related 
costs and anticipated benefits. In the most significant benefit change, 
RSPA reduced its estimate of the number of nonreportable incidents that 
could have benefitted from an EFV installation. Criticisms of its 
estimates on nonreportable incidents led RSPA to conclude that the 
original estimate, over

[[Page 31451]]

143 thousand per year, significantly overstated the number of 
nonreportable incidents whose consequences might be mitigated by EFVs. 
RSPA used a different approach to develop a more reasonable estimate, 
approximately 13 thousand per year, for the final study. This revised 
number of nonreportable incidents is largely responsible for the 
decrease in the present value of the benefits from $21.02-$35.00 per 
service in the draft study to $7.42 per service in the final study.
    In other changes, RSPA revised its cost estimate by using the mid-
point of the cost-range in EFVs. The original estimate looked only at 
the EFV cost to the largest current installers of EFVs, whereas the 
revised estimate considered the EFV cost to all current installers of 
EFVs. RSPA also used newer incident data to develop better estimates of 
the consequences of incidents before and after an EFV installation.
    As a result of RSPA's reexamination of the cost/benefit study, the 
present value of costs changed from the draft study figure of $20.20 
per installed EFV with a bypass to a final study figure of $30.29. In 
addition, in the final study, the present value of costs for an EFV 
with positive shutoff was estimated to be $37.09 per installed EFV.
    The final cost/benefit study found the cost of installing an EFV to 
exceed the benefits by a 4.5:1 ratio. This result, along with 
consideration of other criticisms of a rule requiring installation, 
discussed in more detail below, led RSPA to determine that it would not 
require installation but would require that any EFV installed meet 
certain performance criteria. The final cost/benefit study explains in 
detail how each cost and benefit was calculated. Both the draft and 
final cost/benefit studies examining EFV installation are available in 
the docket.

The Final Rule

    The final rule establishes a new section in the pipeline safety 
regulations, Sec. 192.381, ``Service lines: Excess flow valve 
performance standards.'' For the reasons previously explained, the 
final rule does not require installation of EFVs. In accordance with 49 
U.S.C. 60110, the rule sets performance standards for any EFV that will 
be used in a single-residence gas service line operating continuously 
at not less than 10 psig. The final rule incorporates almost all the 
performance standards that the Joint Commenters recommended, rather 
than those RSPA proposed in the NPRM.
    An EFV will have to be manufactured and tested by the manufacturer 
according to an industry specification or a manufacturer's written 
specification to ensure that the EFV will function properly up to its 
rated maximum operating pressure and at all temperatures expected in 
the service line's operating environment. An EFV, like any other valve, 
will have to comply with subparts B and D of Part 192. The required 
tolerance has been raised so that an EFV will be required to close at, 
or not more than 50 percent above the rated flow, instead of at the 
proposed 10 percent. As commenters requested, an operator will have the 
choice of using an EFV with either a positive shutoff or bypass 
feature. Upon closure an EFV must reduce the gas flow to no more than 5 
percent of the manufacturer's specified minimum flow rate, up to a 
maximum of 20 cubic feet per hour for a bypass-type EFV or 0.4 cubic 
feet per hour for a positive shut off-type EFV. An operator will have 
to mark or otherwise identify the presence of an EFV in the service 
line.
    Several proposed performance requirements have not been adopted. An 
EFV will not have to comply with the requirements of Secs. 192.363 and 
192.365 that apply to other service line valves. Service line capacity 
will not have to exceed the manufacturer's EFV flow rating by 50 
percent. An EFV will not be required to be tested upon installation and 
each time a customer's meter is removed or replaced, or to close 
automatically if the customer's meter, regulator or service valve is 
sheared off. Furthermore, an operator will not be required to verify 
the rated flow or replace an EFV that does not close automatically.
    The final rule recommends that an operator locate an EFV beyond the 
hard surface and as near as practical to the fitting connecting the 
service line to its source of gas supply to ensure that the EFV 
protects the maximum length of service line and to assist in locating 
the EFV. The final rule also recommends that to augment performance 
reliability, an operator not install an EFV where the contaminants in 
the gas stream will cause the valve to malfunction or interfere with 
necessary operation and maintenance activities on the service line, 
such as blowing liquids from the line.

Discussion of Comments

    Although comments were submitted in response to the proposal to 
require installation of EFVs, these comments were also relevant to 
developing a performance standards rule. Many of the comments focussed 
on the performance criteria RSPA included in the proposal.
    General Comments--Except for NTSB, valve manufacturers, and GASAC, 
virtually all of the 140 commenters to the NPRM objected to the 
proposed rule on installation. The major objections were that EFV 
installation should not be federally mandated, that each state pipeline 
authority should be allowed to establish the rules for its state; that 
a positive shutoff EFV should not be required; that testing an EFV 
while in service is unnecessary and overly expensive; that EFV 
installation should be delayed until industry standards are developed; 
and, that the cost/benefit study supporting the proposed rule is 
flawed. The majority of commenters also maintained that EFV 
installation should not be required where contaminants could cause the 
EFV to malfunction and inadvertently shutoff service to the customer.
    Nearly all of the 70 commenters responding to the Notice of 
Reopening Comment Period proposed that RSPA adopt the Joint Commenters' 
recommendations on performance language because the recommended 
language was less objectionable than the NPRM's proposed language. The 
commenters also favored giving an operator the option to install either 
a bypass or positive shutoff EFV. Overall, because of concerns about 
EFV reliability, gas distribution operators favored waiting until 
industry standards are developed and accepted before requiring 
installation of EFVs. Many commenters restated their objection to the 
findings of the cost/benefit study.
    Six large operators operating at least 9 million service lines (18 
percent of all U.S. service lines) opposed both the NPRM's proposal and 
the Joint Commenters' recommendations. The operators' major objections 
were that the cost/benefit study grossly overstated benefits, that 
industry standards are needed because EFVs do not operate reliably, and 
that costs to remove EFVs after a malfunction are high.
    Comments about the cost/benefit study have previously been 
discussed. Other general comments are discussed below, as well as 
specific comments about each RSPA-proposed performance standard and the 
associated Joint Commenters' recommendation. To avoid repetition, 
similar comments are discussed in only one section.

Discussion on State vs. Federal Mandate

    Comments--NAPSR expressed opposition to any federal mandate to 
install EFVs, arguing that any such regulatory requirements should be 
at the state level. On two occasions NARUC passed resolutions proposing 
that any requirement for EFVs be determined by

[[Page 31452]]

the individual state pipeline safety agencies. The NARUC Subcommittee 
for Pipeline Safety polled the state regulatory agencies, gathered 
data, and prepared a report of its findings. NARUC found that only two 
states, Massachusetts and New York, favored a federal mandate to 
install EFVs.
    Six major operators (three operating in California) opposed any 
federal requirement to install EFVs, arguing that states should be 
allowed to determine the need for EFVs based on state- developed 
criteria.
    Response--Because of RSPA's decision not to issue a rule requiring 
the installation of EFVs, each state will be able to determine if it 
should require such installation based on circumstances unique to that 
state.

Industry Standards

    In the absence of standards by an industry-sponsored safety 
standards committee, RSPA proposed several requirements for the 
manufacture and operation of any EFV that would be installed in a 
single-residence gas service line. The Joint Commenters' recommendation 
also included performance standards for single- residence gas service 
lines.
    Comments on NPRM--Many commenters said RSPA should not issue a 
final rule until industry manufacturing and performance safety 
standards are prepared and adopted. The TPSSC recommended that RSPA 
initiate the development of standards by The American National 
Standards Institute (ANSI), American Society of Testing Materials 
(ASTM), or other nationally recognized and accredited organization for 
the manufacture, testing, and operation of EFVs. The TPSSC further 
recommended that when such standards are enacted, RSPA should issue an 
NPRM for EFVs incorporating such standards for TPSSC review. The Gas 
Piping Technology Committee (GPTC) commented that its ANSI/GPTC Z380 
committee was developing performance, operating, and installation 
guidelines for EFVs. GPTC said guidance will be offered on choosing 
operating pressure ranges, flow rates, bleed-by, and reset 
characteristics, length and diameter of service piping, inline 
contaminants, purging procedures, joining methods, and service line 
locations.
    Comments to Notice of Reopening Comment Period--Many commenters 
said RSPA should take no final action until industry standards are 
available because standards would assure EFV reliability. Many others 
said RSPA should issue a final rule but grant a one year delay in 
implementation to give the industry committees time to complete 
manufacturing and operational standards. Several commenters said the 
ASTM F17 committee is preparing testing standards and the ANSI/GPTC 
Z380 committee is preparing guidelines that should be completed in 
1995.
    Response--RSPA agrees that to achieve performance reliability and 
the desired safety benefits, specifications are necessary to ensure 
uniformity among EFVs installed in service lines. Because the NPRM 
proposing required installation only sought comment on performance 
standards applicable to EFVs installed in single-residence service 
lines, this final rule limits EFV performance standards to that 
application. Once industry standards are developed for EFVs used in 
other applications, such as multiple residences and commercial 
enterprises, RSPA will consider seeking comment on proposed performance 
standards for those applications.
    The final rule requires that when an EFV is installed in a single 
residence service line, the EFV must be manufactured and tested by the 
manufacturer according to an industry specification, or to a 
manufacturer's written specification to ensure the EFV performs 
specified minimum functions. These specifications will ensure that an 
EFV functions properly up to the maximum operating pressure at which it 
is rated and at all temperatures reasonably expected in the service 
line's operating environment. These specifications will further ensure 
that an EFV is sized to close within 50 percent of the rated closure 
rate, to reduce gas flow upon closure to specified rates, and to not 
close when the pressure and flow rates are less than the manufacturer's 
specified minimums.
    In addition, an EFV must comply with the general requirements of 
Subparts B and D of part 192. While subparts B and D do not include 
operational requirements specific to an EFV, they do include general 
material and design standards applicable to any valve in a pipeline 
system.
    Many commenters, including several industry committees, indicated 
that EFV standards are forthcoming. However, until industry finalizes 
EFV standards, the requirement that an EFV perform specified functions 
according to a manufacturer's written specifications will ensure that 
an EFV performs reliably and safely. Moreover, final industry 
performance specifications are likely to be similar to manufacturers' 
specifications, because valve manufacturers are often members of the 
industry organizations that develop such specifications.
    Proposed Section 192.381(a)--(regarding Secs. 192.363 and 192.365 
gas pipeline valve requirements)--RSPA proposed in the NPRM that EFVs 
must comply with the requirements of Secs. 192.363 and 192.365. These 
existing sections establish requirements for all valves in gas service 
lines.
    Comments--Several commenters stated that Secs. 192.363 and 192.365 
should not apply to EFVs. Commenters pointed out that these 
requirements apply to the design of service line manual shut-off valves 
and would conflict with the proposed EFV requirements. For example, 
commenters noted that the Sec. 192.365(c) requirement to locate valves 
in a covered durable box or standpipe is intended to allow for ready 
operation of a service line manual shut-off valve. Therefore, it would 
be unnecessary and costly to apply this requirement to an EFV, which is 
an automatic valve not requiring access for manual operation.
    Response--After further study, RSPA agrees that valve requirements 
concerning the use of a durable box or standpipe do not apply to EFVs, 
and the other requirements of Secs. 192.363 and 192.365 apply only to 
manual shut-off type valves, not EFVs. Accordingly, the proposed 
requirement that EFVs comply with Secs. 192.363 and 192.365 has not 
been adopted.
    Proposed Section 192.381(a)--(10 psig requirement)-RSPA proposed 
that an EFV be installed on each newly installed or replaced single 
residence service line that operates at a pressure not less than 10 
psig.
    Comments--Many commenters to both the NPRM and the Notice of 
Reopening Comment Period requested clarification of the 10 psig 
threshold. Many commenters asked if the requirement would apply if 
pressure in the pipeline system drops below 10 psig at any time during 
the year.
    Response--RSPA is not requiring operators to install EFVs on any 
single-residence service line, whatever its operating pressure. 
However, RSPA does not want an EFV, if installed, to cause a loss in 
service, especially at a time when the service is most needed by the 
consumer, such as during the winter heating season. Thus, the 
performance standards have been established for EFVs that are installed 
on a service line that operates at or above 10 psig continuously during 
the year. Setting the performance standards at this threshold is 
influenced by two of the largest users of EFVs who, as standard 
practice, limit EFV installation to service lines in systems where 
service line inlet pressure does not drop below 10 psig during the 
year.

[[Page 31453]]

    Because service line pressure will most likely be at its lowest 
level during the coldest weather, especially in colder climates, an 
operator should consider the pressure drop in the service line due to 
the restriction of gas flow caused by an EFV. If pressure drop is 
considered, an EFV should not cause a reduction in safety or loss of 
service in any service line.
    Proposed Section 192.381(a)--(replaced service lines)--RSPA 
proposed that EFVs be installed on certain new and replaced service 
lines.
    Response--This proposal is no longer relevant since EFV 
installation is not being required.
    Proposed Section 192.381(b)(1)--(installation)--RSPA proposed in 
the NPRM that an EFV be installed as close to the main or transmission 
line as practicable. The Joint Commenters recommended installation in 
or as near as practicable to the service line fitting connecting the 
service line to its gas supply.
    Comments--Many commenters suggested RSPA remove any reference to 
transmission lines in the rule. Several commenters said EFVs are not 
available that will withstand transmission line pressures. Others 
stated that the statutory mandate was intended to apply only to 
distribution systems. The TPSSC voted 7 to 4 that all references to 
transmission lines be dropped from the proposed rule.
    A few commenters objected to what they thought was the proposed 
requirement to install EFVs immediately downstream of the service-to-
main connection when the line serves more than one residence (branch 
service). Other commenters were concerned that the proposed rule would 
require EFV installation below hard surfaces such as asphalt or 
concrete, making installation very costly.
    Response--In the NPRM, RSPA intended that all new and replaced 
service lines, whether from a main or transmission line, where the 
source of gas supply consistently operates above 10 psig, be required 
to have an EFV installed. The reference to ``main'' and 
``transmission'' lines was intended to cover farm taps, as farm taps 
are also subject to the type of incident that could benefit from an 
EFV. The final rule deletes the reference to ``main'' and 
``transmission'' and sets performance standards for EFVs installed on 
single-residence gas service lines. By referring to ``service'' line, 
RSPA intends for the standards to apply if an EFV is installed on a 
farm tap. A farm tap operates as a service line when a local 
distribution company operates a metered farm tap on a transmission line 
delivering gas to a farmer or other landowner. Accordingly, although 
the rule does not require installation on any single-residence service 
line, an EFV that meets the required performance standards can be 
installed on a service line from a main or a branch off a transmission 
line.
    RSPA never intended that an EFV serve more than one family 
residence. RSPA recognizes that an EFV would be difficult to size when 
the gas supply is serving multiple residences because of widely varying 
gas volume through the EFV. Because of this difficulty, the performance 
standards in this final rule are limited to EFVs that are installed on 
single-residence service lines.
    RSPA agrees that removing an EFV under a hard surface would be 
overly expensive if an EFV failed to function. Therefore, RSPA 
recommends that an EFV be located beyond the hard surface and as near 
as practical to the fitting connecting the service line to its source 
of gas supply.
    Proposed Section 192.381(b)(2)--(Section 192 Subparts B & D)--As 
noted above, the NPRM proposed and the Joint Commenters recommended 
that EFVs meet the applicable requirements of subparts B and D of part 
192.
    Comments--No substantive comments were received on this proposal.
    Response--Subpart B establishes minimum requirements for selection 
and qualification of materials to be used in pipelines. Subpart D 
prescribes minimum requirements for the design and installation of 
pipeline components and facilities. Since these requirements are 
general performance requirements that apply to all valves, they are 
included in the performance requirements applicable to EFVs.
    Proposed Section 192.381(b)(3)--(bypass)--RSPA proposed that an EFV 
be designed to prevent pressure equalization across the EFV after the 
EFV closes, thereby prohibiting an operator from installing an EFV with 
a bypass feature. The bypass feature allows pressure to equalize and 
the EFV to automatically reopen after closure because it allows a small 
amount of gas to pass through the EFV. In contrast, a positive shutoff 
feature allows only minute amounts of gas to pass through the EFV after 
it closes, and requires backpressuring downstream to reset the EFV. The 
Joint Commenters' recommendation would allow either type of EFV.
    In the Notice of Reopening Comment Period, RSPA sought comment on 
the safety of using EFVs with or without the bypass feature and gave 
two examples, provided by two large local distribution operators, of 
potential dangers that might be caused by the bypass feature. RSPA also 
asked for comments on the conditions under which automatically 
resetting EFVs should or should not be required in residential service 
lines and on the linkage between the bypass feature and unauthorized 
repairs to damaged service lines.
    Comments to NPRM--Many commented on the proposal prohibiting the 
use of EFVs with a bypass feature. Commenters, including several at the 
public meeting, were virtually unanimous in favor of an operator having 
the option to select an EFV with either the bypass or positive shutoff 
feature. Similarly, the TPSSC voted 9 to 2 in favor of an operator 
having this option.
    Various reasons were given for not prohibiting the installation of 
bypass EFVs. Several commenters, including an industry association, 
complained that RSPA proposed the positive shutoff requirement without 
sufficient justification in the cost/benefit study. One commenter said 
that additional costs of at least $250 per utility crew would be 
incurred to provide backpressure downstream of the EFV to equalize the 
pressure and reset the valve. This commenter said these services would 
necessitate extra equipment, including a compressed natural gas tank or 
portable natural gas compressor, and additional piping, fittings, and 
hoses. Other commenters mentioned additional hazards to personnel in 
hauling and connecting compressed natural gas. Another commenter was 
concerned with customer inconvenience because a service call would be 
necessary to backpressure the EFV, delaying restoration of service.
    Many commenters argued that bypass-type EFVs do not pose a 
significant safety risk. Commenters maintained that operators that 
regularly install EFVs have had no incidents resulting from use of 
bypass-type EFVs. Three of the largest voluntary users of EFVs (with 
over 300,000 EFVs in service) commented that their data did not show an 
incident having occurred due to a bypass- type EFV. An EFV manufacturer 
commented that it has no knowledge of bypass gas ever contributing to a 
natural gas incident. NTSB and many operators echoed these assurances.
    Several commenters, including EFV users, said RSPA's concern that 
the bypass feature would allow irresponsible excavators to make repairs 
is unfounded. A few commenters said that positive shutoff EFVs would 
cause more safety problems than bypass-type EFVs because an excavator 
could sever a service line unknowingly if the

[[Page 31454]]

positive shutoff were to completely stop the gas flow and any released 
odor from reaching the atmosphere. Conversely, these commenters argued 
that a failed service line with a bypass would continuously release gas 
and leave a readily detectable odor. Commenters noted other potential 
problems with positive shutoff EFVs. For example, a commenter in Alaska 
pointed out that an earthquake in the winter could cause EFVs to engage 
and, if positive shutoff EFVs were used, each would have to be 
backpressured and each customer's appliance re-lighted. During an 
Alaskan winter this could take days.
    The Gas Research Institute (GRI) stated that its tests of EFV 
models showed all the tested models were affected by pressure surges of 
5 psi or more and that opening, closing, or throttling a main line 
valve could activate an EFV, causing a false closure. The research 
organization said RSPA could infer from these results that the use of 
EFVs without the bypass could cause extended distribution service 
outages. GRI further stated that it knows of no reports of bypass flow 
in an EFV having led to or increased the severity of an accident.
    GASAC commented that RSPA should allow each operator to determine 
the type of valves for its system. Other commenters echoed this 
statement. Even among those operators opposed to a mandatary rule, most 
said that if a rule were issued, the choice of which type of EFV to use 
should be left to the operator.
    Comments on the Joint Commenters' Recommendation - Many commenters 
supported the Joint Commenters' recommendation to allow the use of a 
bypass-type EFV. Many commenters said it is not appropriate to depend 
on an EFV's design to prevent unauthorized repairs. Rather, 
unauthorized repairs should be controlled by stiffer penalties and 
better enforcement of damage prevention laws. These commenters 
maintained that EFVs are used to provide safety when a service line is 
severed, and should not be expected to perform functions beyond their 
intended purpose.
    Many commenters said excavators who damage service lines may make 
unauthorized repairs regardless of whether a bypass-type EFV, a 
positive shutoff EFV, or no EFV is installed. RSPA recognizes the 
validity of this statement and that EFVs with either feature are not 
likely to have a substantial effect in either reducing or increasing 
the frequency of unauthorized repairs on a broken service line.
    To dispel RSPA's concern about the potential danger of bypass- type 
EFVs and gas discharge into a residence, an operator explained that 
since natural gas is only about 0.6 times the density of air, any raw 
gas passing through a vented appliance would exhaust to the atmosphere 
through the chimney. The operator concluded that household gas ranges 
(or space heaters) without burner safety pilots are the only paths for 
raw gas to disperse through a building. The operator cited a recent 
study by NOVA, a Canadian chemical and pipeline company, that 
demonstrated that a rate of raw gas buildup in a small residence (300 
square feet) would have to be about 60 cubic feet per hour to reach an 
ignitable level in five hours. This allows a five hour period for 
someone to discover the gas release before the ignitable level is 
reached. A bypass-type EFV allows 20 cubic feet of gas per hour. 
Therefore, natural gas that is passing through an EFV with a bypass 
would take several hours to accumulate to the ignitable range in a 
building.
    Response--RSPA has been concerned that excavators could repair a 
service line break equipped with an EFV with a bypass feature, the EFV 
would automatically reset, and service would be restored without the 
operator knowing that the line had been damaged. Consequently, gas 
could then pass into and accumulate in a residence where the pilot 
light on a gas appliance had been extinguished during the service line 
break.
    RSPA was also concerned that restoration of gas service with 
unvented appliances would cause a rapid buildup of the gas/air mixture 
to an ignitable level. Commenters have posed circumstances under which 
such a buildup could occur. However, in response to its questions about 
this problem, RSPA did not receive any information that such an 
incident has actually occurred. Furthermore, an EFV manufacturer and 
AGA have assured RSPA that bypass-type EFVs operate properly to avoid 
unintended gas buildup within a building. An operator with 20,000 
installed bypass-type EFVs stated that bypass gas from a tripped EFV 
had never caused or contributed to an unsafe situation on its system. 
Other operators made comparable statements. The NOVA study, described 
above, further allays RSPA's concern. Therefore, based on the record in 
this rulemaking, RSPA accepts the premise that EFVs with a bypass 
feature are safe.
    RSPA also finds acceptable the Joint Commenters' recommendation to 
limit gas flow to 20 cubic feet per hour for bypass-type EFVs and to 
0.4 cubic feet per hour for positive shutoff-type EFVs. Because EFVs 
with positive shutoff features were proposed in the NPRM, RSPA did not 
propose EFV flow limits. However, RSPA agrees that the limits 
recommended by the Joint Commenters are reasonable and feasible design 
requirements.
    Accordingly, the final rule allows either bypass or positive 
shutoff EFVs. Closure flow rates will be limited to 20 cubic feet per 
hour for the bypass-type EFV and 0.4 cubic feet per hour for the 
positive shutoff EFV.
    Proposed Section 192.381(b)(4)--(installation testing)--RSPA 
proposed that upon original installation of an EFV and each time the 
meter is removed or replaced, the EFV be tested to determine if it 
closes automatically. The Joint Commenters' recommendation deleted the 
requirement.
    Comments--All 37 commenters on this proposed requirement asked that 
it be deleted. Most commenters stated that the test would require that 
the service line be disconnected from the meter set, the service valve 
at the meter opened, and gas vented to the atmosphere to trip the EFV. 
Many commenters said that venting of the gas near the residence, or 
inside the residence when the meter is indoors, would be hazardous and 
would needlessly release methane into the atmosphere contrary to the 
goals of the Clean Air Act.
    An EFV user stated that it does not test the EFV when replacing 
meters. This commenter stated that it replaces one-tenth of its meters 
annually and provided RSPA a summary of the steps involved in testing 
an EFV when a meter is replaced on an existing service. This commenter 
further stated it would take a two person crew a full day to test an 
EFV, resulting in substantial cost with no corresponding benefit. The 
American Public Gas Association (APGA) commented that the proposed 
testing would add significantly to the costs of using EFVs with no 
corresponding safety benefits and noted that these costs were not 
included in the cost/benefit analysis.
    Several other commenters also noted that this proposed requirement 
had not been covered in the cost/benefit analysis and provided data on 
the costs that would be incurred for such tests. AGA estimated that 3 
million services have meters removed each year, so that the tests could 
cost $100 million per year, doubling RSPA's estimated installation cost 
of $20 per EFV (with bypass feature). These same commenters contended 
that testing positive shutoff EFVs would cost even more.
    AGA and other commenters concluded that such tests would require 
removing the service regulator or

[[Page 31455]]

installing a fitting to allow gas to be vented upstream of the service 
regulator because the flow of gas passing through a service regulator 
may be too small to cause the EFV to trip. These commenters said that 
such a fitting would invite a resident to bypass the meter and steal 
gas.
    The TPSSC voted 8 to 2 that no in-service testing of an EFV be 
required.
    Response--Based on the comments about problems and costs of 
installation testing, the final rule will not require an operator to 
test the EFV when the EFV is installed or when the meter is removed or 
replaced. However, the requirement that the EFV must be manufactured 
and tested to an industry specification or manufacturer's written 
specification to ensure that the EFV functions properly up to the rated 
maximum operating pressure will certainly require random sample testing 
at the manufacturer's plant. Such sample testing is routinely conducted 
for all other valves in accordance with manufacturing standards.
    Proposed Section 192.381(b)(5)--(automatic closure)--RSPA proposed 
that an EFV must close automatically if the service line is severed or 
if the customer's meter, regulator, or service valve is sheared off. 
The Joint Commenters' recommendation did not include such a 
requirement.
    Comments--All seventeen commenters on this proposed requirement 
argued that it should be deleted. Most commenters stated that operators 
cannot guarantee that an EFV will perform as designed and warranted by 
the manufacturer. One commenter said that it would be difficult to 
comply with such a requirement because EFVs often fail to activate (due 
to fluid friction) in longer service line lengths of \1/2\-inch pipe. 
Also, even if the meter set is sheared off, the flow rate may not 
exceed the EFV activation flow rate because the pipe may be squeezed 
off at the point where it is sheared, or because there are other 
restrictions in the line.
    One EFV user stated that costs for assuring that an EFV closes 
automatically would approach $1,000 per installation. This commenter 
reasoned that an EFV is intended to help reduce the effects of dig-ins 
on a service line in the area of the street, where most excavation 
takes place, and requiring the EFV to do more than intended will 
increase costs.
    The TPSSC voted 7 to 3 that the proposed requirement be changed so 
that an EFV ``be designed to close automatically if the service line is 
ruptured downstream of the valve.''
    Response--RSPA agrees with the commenters that flow rate may not 
always exceed an EFV's activation flow rate because a long service line 
could cause excessive pressure drop, or a line could be squeezed off at 
the point where it is sheared, or there could be other restrictions in 
the line. Therefore, RSPA is not including proposed Sec. 192.381(b)(5) 
in the performance standards. However, the final rule (Sec. 192.381(c)) 
requires that an EFV be manufactured according to an industry 
specification or manufacturer's written specification that will 
establish shutoff requirements for conditions comparable to a service 
line being severed or a meter set being sheared off.
    Proposed Section 192.381(b)(6)--(sizing)--RSPA proposed that an EFV 
be sized to close within 10 percent of the rated flow specified by the 
manufacturer. The Joint Commenters recommended a closure rate not less, 
and not more than 50 percent higher, than the manufacturer's specified 
closure flow rate.
    Comments to NPRM--The 32 commenters objected to this requirement. 
Most commenters suggested that the proposed 10 percent tolerance be 
raised to 50 percent because EFVs are not precision instruments. Some 
commenters suggested a 25 percent tolerance. Most commenters said that 
EFVs with 10 percent tolerance are not commercially available and would 
be significantly more expensive. GASAC also opposed the requirement as 
excessive.
    AGA provided exhaustive information showing that EFVs with a 10 
percent tolerance are not commercially available and may not be 
possible to mass produce. AGA suggested a 50 percent tolerance and 
cited a Gas Research Institute (GRI) study regarding EFV performance 
repeatability. In 1985, GRI tested seven EFV models and found that 
closure flows of a single copy were repeatable within a range of 6.4 
percent to 20.8 percent, whereas closure flows between two arbitrary 
copies of the EFVs were repeatable within the range of 15.4 percent and 
87.9 percent. None of these models would have met the RSPA proposed 
requirements. AGA provided an EFV manufacturer's graphs showing that 
none of the currently available EFVs tested by that manufacturer closed 
within 10 percent of the rated closure.
    Comments on Joint Commenters' recommendation--A member of the Joint 
Commenters said its analysis of service ruptures found that EFVs could 
close as much as 50 percent over specified closure flow and still 
reliably close in the type of accident EFVs are meant to address. Three 
other commenters agreed with the Joint Commenters' recommendation.
    The TPSSC voted 7 to 4 that the rule specify that an EFV must close 
no lower than its rated flow and not more than 50 percent above rated 
closure flow.
    Response--Although no EFV is currently available at an acceptable 
cost that will conform to a 10 percent tolerance, RSPA believes that 
distribution operators must have a specified closure range for an EFV 
that is reliable. The requirement that an EFV activate at, or 50 
percent above, a specified flow level provides an acceptable closure 
range in accordance with currently available EFVs. Accordingly, RSPA 
will require an EFV be sized to close at or 50 percent above the rated 
closure flow rate specified by the manufacturer.
    Proposed Section 192.381(c)--(flow rate verification)--RSPA 
proposed that the operator verify the manufacturer's rated flow for the 
EFV by testing at a pressure of 10 psig for the gas to be transported 
in the service line. The Joint Commenters recommended that the 
manufacturer certify the EFV meets the manufacturer's written 
performance specifications, rather than place this responsibility on 
the operator.
    Comments to NPRM--Thirty six commenters responded to RSPA's 
proposed requirement. Virtually all commenters objected to any operator 
responsibility for testing and suggested the requirement be deleted. 
Most commenters contended that operators cannot guarantee the 
performance of an EFV, but should be able to rely on the manufacturer 
to certify that EFVs meet the applicable standards--the approach 
allowed for other valves used in gas distribution systems. An EFV 
manufacturer also agreed that it should be the manufacturer's 
responsibility to test and certify EFVs. Most commenters stated that 
the proposed requirement would significantly increase an operator's 
costs.
    Comments on Joint Commenters' recommendation--An industry 
association agreed with the recommendation to allow an operator to rely 
on the manufacturer's certification that EFVs meet performance 
standards rather than have the operator test each EFV. The association 
pointed out that RSPA allows such a procedure under Sec. 192.145.
    Response--RSPA agrees with the commenters that the flow rate 
verification test should be an EFV manufacturer's responsibility, not 
the operator's. Thus, the final rule requires that an EFV be 
manufactured and tested by the manufacturer according to an industry 
specification, or

[[Page 31456]]

manufacturer's written specification to ensure that each valve will 
perform specified minimum functions. This requirement should lead to a 
random EFV testing program by the manufacturer, similar to testing for 
other system valves. Currently, certain valves (cast iron and plastic) 
are installed that meet the specified manufacturing tests in 
Sec. 192.145. All other valves must be manufactured according to 
specifications in American Petroleum Institute (API) Standard 6D, which 
also requires random testing by the manufacturer.
    Proposed Section 192.381(d)--(replacement)--RSPA proposed that if 
an EFV does not close automatically during installation testing or when 
the service line is severed, it must be replaced with an EFV that 
closes as required. The Joint Commenters' approach would remove any 
requirement to assure that an EFV closes after installation.
    Comments--None of those commenting on RSPA's proposal was entirely 
satisfied with it. Seven commenters suggested changes that included 
permitting the operator the option to repair or replace an EFV that 
doesn't close. These commenters further proposed exempting a location 
from the installation requirement after two EFVs do not perform 
properly at that location.
    One operator questioned what constitutes satisfactory closure by 
explaining that minor accumulations of dust and dirt can interfere with 
an absolute 100 percent shutoff. This commenter said that RSPA should 
conduct additional studies to ascertain what long-term performance 
characteristics can be expected and include acceptable criteria in the 
rulemaking.
    Eight commenters said the requirement was not needed or questioned 
the apparent intent to require the operator to keep replacing an EFV 
until one performs as required. Several said that the requirement 
assumed that an EFV's failure to close is always the valve's fault. 
Commenters explained that many factors influence the operation or 
performance of an EFV, including changes in operating pressures and the 
type of gaseous mixtures flowing through the service line. They 
suggested the practical approach would be to allow the utility to 
repair and replace an EFV at its own discretion as it does with other 
valves in its system.
    Response--RSPA's proposed requirement that an operator replace an 
installed EFV if it fails during installation testing or during a 
service line break, is no longer applicable since on-site testing and 
mandatory EFV installation are not being required in this final rule. 
Instead, an EFV must be manufactured and tested by the manufacturer 
according to an industry specification or manufacturer's written 
specification to ensure that the valve will function properly. 
Furthermore, replacement or removal of a defective EFV will be left to 
agreement between the customer and operator.
    Section 192.381(e)--(manufacturing specifications)--RSPA proposed 
that each EFV must be manufactured in accordance with written 
specifications that assure the EFV meets the manufacturer's published 
pressure and flow rate criteria. The Joint Commenters recommended that, 
instead, an EFV be manufactured and tested by the manufacturer 
according to a written specification to ensure that the EFV will 
function properly up to the maximum rated operating pressure and at all 
temperatures reasonably expected. The Joint Commenters further 
recommended that an EFV not close when pressures are below the 
manufacturer's minimum pressure.
    Comments--Fourteen of the fifteen commenters responding to RSPA's 
proposed requirement were dissatisfied with the wording and recommended 
changes. These commenters stated that this provision appeared to shift 
responsibility for quality assurance from the manufacturer to the gas 
distribution operator who cannot assure that the manufacturer will 
produce valves meeting the manufacturer's published pressure and flow 
rate criteria. Commenters further stated that because of liability 
concerns there should be an industry EFV standard by which the valves 
should be manufactured. APGA also argued that manufacturers, not gas 
distribution operators, should be responsible for assuring that EFVs 
meet the necessary performance criteria.
    Response--RSPA agrees that the proposed requirement was unclear as 
to who would be responsible for assuring that an EFV meets the 
specified performance requirements. Accordingly, the final rule 
clarifies that an EFV will have to be manufactured and tested by the 
manufacturer according to an industry specification or manufacturer's 
written specification to ensure that each valve meets the specified 
minimum performance standards.
    Proposed Section Sec. 192.381(f)--(service line capacity)--RSPA 
proposed that service line capacity must exceed the EFV manufacturer's 
flow rating by 50 percent. The Joint Commenters' approach did not 
include a similar requirement.
    Comments on NPRM--Thirty three commenters responded to this 
proposed requirement. Five commenters said that maintaining a flow rate 
at least 50 percent over the rating of the EFV would severely restrict 
an operator and increase costs. These commenters explained that such a 
high flow rate would, in many cases, require the installation of 
service lines larger in diameter than required for a customer's load 
and also preclude the insertion of plastic tubing. These persons 
recommended reducing the flow rate margin to 25 percent.
    Most commenters opposed establishing arbitrary excess flow 
capacity. These commenters stated that the sizing of service lines is 
the operator's responsibility and that many factors must be considered, 
such as costs, current and future loads, the possibility of future 
insertions, and future maintenance requirements.
    Response--RSPA agrees that a requirement to design a service line 
with excess capacity is not necessary for an EFV to function properly 
and would add unnecessary expense. Thus, the final rule does not 
require that service line capacity exceed the EFV manufacturer's flow 
rating by 50 percent. This approach is consistent with Part 192, which 
does not require installation of service lines larger than required to 
meet the customer's load.
    Proposed Section 192.381(g)--(Marking)--RSPA proposed that each 
service line with an EFV be physically marked or labeled in the field, 
so that the label would be readily visible to gas company employees.
    Comments on NPRM--Twelve commenters said that requiring service 
lines with EFVs to be identified is unnecessary and is of little 
benefit. One commenter, currently using EFVs and marking those service 
lines, said it does not believe that marking should be required. 
Several commenters stated that marking service lines is futile due to 
customers painting the meter set, weather deterioration, and vandalism. 
A few commenters suggested that the operator have the option to mark or 
record the location of these valves. However, eight commenters 
supported the requirement, saying it is a good safety practice for gas 
company operator personnel, when arriving at a residence, to know if an 
EFV is installed in that service line.
    Comments on Joint Commenters' Recommendation--The Joint Commenters' 
recommendation did not include a requirement to mark services in the 
field. An industry association supported the Joint Commenters' approach 
and further recommended that

[[Page 31457]]

the operator be allowed the option to mark services in the field or 
record EFV installation on its maps and records.
    Response--RSPA believes it is helpful for operating personnel to 
know if an EFV is installed in a service line. In a service outage or 
emergency, service personnel arriving at a residence might respond 
differently depending on whether or not an EFV is installed. For 
example, if service personnel find that a service line has been severed 
and the line is marked or otherwise identified as having an EFV, 
service personnel should recognize that the small amount of gas 
escaping from the severed line is from an EFV with a bypass feature and 
not from a pinched service line that could suddenly release a hazardous 
flow of gas. With this knowledge, service personnel can initiate 
correct repair procedures.
    Accordingly, the rule will require that an operator must mark or 
otherwise identify the presence of an EFV in the service line.
    Proposed Section 192.381(h)--(Contaminants)--RSPA proposed that EFV 
installation not be required on a service line where the operator can 
demonstrate that contamination in the gas stream will cause an EFV to 
malfunction. The Joint Commenters' approach eased the operator's burden 
of proof by allowing the operator to document, rather than demonstrate, 
an unsatisfactory level of contamination.
    The Joint Commenters also recommended that EFV installation not be 
required where the EFV would interfere with operation and maintenance 
activities, such as blowing liquids from the line.
    Comments on NPRM--Twenty-four commenters supported the proposal to 
except EFV installation where prior experience indicates contaminants 
will cause a malfunction. Several commenters stated, however, that it 
is unclear how an operator could make such a demonstration. NTSB said 
RSPA should state the requirements necessary to claim the exemption. 
Several commenters said they hoped that an operator would not have to 
install an EFV and wait for it to fail before being able to demonstrate 
that contaminants should preclude installation. Two commenters argued 
that if an operator has experience with clogging of valves, regulators, 
or meters from liquids or solids in certain areas of its system, such 
experience should be sufficient to demonstrate that an EFV should not 
be installed on that part of the system.
    An EFV manufacturer agreed that an EFV should not be installed 
where contaminants would interfere with the proper operation of an EFV, 
but based on its experience felt it unlikely that many systems have 
sufficient contaminants to cause an EFV to malfunction. GASAC commented 
that requests for an exemption should be subject to public disclosure 
and a formal review process to prevent unwarranted exemptions.
    Comments on Joint Commenters' recommendation--AGA argued that the 
operator should determine whether to use EFVs in contaminated areas. 
AGA said a company might cite previous experience with service lines 
plugging with liquids or solids, plugging of other valves or service 
regulators, or knowledge of liquids or solid debris in certain parts of 
the system to justify not installing EFVs.
    Another commenter said that iron oxide rouge from steel pipe mixed 
with tiny amounts of compressor fluids forms a sticky residue and 
prevented early model EFVs from successfully resetting following 
closure. The commenter said it is likely that no EFV on the market 
today is robust enough to withstand such contaminants and operate 
properly for the minimum expected life of 50 years estimated in the 
NPRM.
    Response--RSPA agrees that an EFV is not recommended on a service 
line where the operator has prior experience with contaminants in the 
gas stream that could interfere with the EFV, cause loss of service to 
a residence, or cause an operator to incur undue expense in removing an 
inoperative EFV. An operator should, based on its previous history of 
service line or equipment problems from contaminants, decide whether it 
is appropriate to install an EFV. An operator should also consider if 
an EFV installed on a service line could interfere with the operator's 
operation and maintenance procedures.

Regulatory Notices and Analyses

Executive Order 12866 and DOT Regulatory Policies and Procedures

    This final rule is a significant regulatory action under Executive 
Order 12866. Therefore, it was reviewed by the Office of Management and 
Budget. In addition, the final rule is significant under DOT's 
regulatory policies and procedures (44 FR 11034; February 26, 1979) 
because it concerns a matter of substantial interest to the public and 
Congress.

Cost/Benefit Analysis (EFV--Performance Standards)

    Since the final rule does not require mandatory installation of 
EFVs, the performance requirements of this rule will not impact gas 
distribution systems not currently installing EFVs unless they begin 
installing EFVs. This rule will impact manufacturers of EFVs. As 
previously mentioned, OPS will be initiating a separate rulemaking to 
propose that customers be notified that EFVs are available for 
installation and will be installed at customer expense. This means that 
all gas distribution systems may soon be installing EFVs, and, thus, 
may be impacted by the new EFV performance standards.
    The new EFV performance standards will help ensure that gas 
distribution companies that currently install EFVs, as well as those 
that begin to install EFVs on their own or because of a new 
notification rule, properly install these EFVs. Furthermore, these 
standards, by helping to ensure that newly installed EFVs are 
manufactured to function properly (e.g., close when they are supposed 
to and not close when they are not supposed to), will reduce the cost 
of improper closure to both gas distribution system operators and the 
general public. The standards will also help keep substandard valves 
from entering the marketplace, thereby providing some assurance of 
reliability to both operators and customers. As a further result of 
these standards, reliable EFVs installed on compatible service lines 
will help mitigate the consequences of incidents on service lines.
    The cost/benefit study accompanying this rule estimates and 
compares the benefits and costs of the EFV performance standards to 
determine whether the standards, taken as a whole, would be cost 
beneficial. This study estimates the expected benefits and costs of 
installing one EFV and uses these estimates to calculate a benefit/cost 
ratio. This approach yields the same benefit/cost ratio as an approach 
considers the number of EFVs installed in each year, but is less 
complicated and cumbersome, since it does not require the estimation of 
(1) the number of services expected to be renewed each year, (2) the 
number of new services expected to be installed each year, and (3) the 
number of existing services that will be discontinued each year.
    The primary sources of EFV data used in the analysis were (1) the 
written submissions to the Docket for this rulemaking made by gas 
distribution companies, EFV manufacturers, and other interested parties 
and (2) direct contacts with gas distribution companies, EFV 
manufacturers, and other interested parties.
    The pipeline incident data used in this analysis was taken 
primarily from the incident and annual report submissions made to OPS 
by gas distribution companies. These

[[Page 31458]]

submissions are required under the Federal pipeline safety regulations.
    All dollar figures in the study are given in nominal dollars, 
unless otherwise indicated. Where deflation of nominal dollar figures 
has been performed, the Producer Price Index, All Commodities, with 
1993 as the base, has been used.
    As summarized below, benefits, costs, and net benefits were 
developed for (1) the standards for EFV installation, (2) marking 
requirements, and (3) the performance requirements. The complete 
Benefit/Cost Analysis for EFV Performance Standards, dated August 1995, 
is available in the Docket.

Standards for EFV Installation

    The final rule requires that an EFV installed on a single-family 
residential gas service that always operates at 10 psig or greater (1) 
must be rated by the manufacturer for use at the pressure and flow rate 
anticipated on the service line and (2) must meet the applicable 
requirements of Subparts B and D of Part 192. The final rule also 
recommends that an installed EFV be placed as near as practical to the 
main. Although this rule specifies standards for EFV installation, the 
installation of EFVs is not mandatory. However, if an EFV is installed, 
the regulatory standards will help ensure the EFV performs as expected 
and protects the maximum length of the most vulnerable portion of a 
service line.
    The standards for EFV installation appear to be consistent with 
current industry practice. Consequently, the benefits, costs, and net 
benefits of the requirements are all expected to be $0 per EFV per 
year.

Marking Requirements

    The new marking requirement will enable gas distribution system 
operating and service personnel to know if a service line has an EFV 
installed when responding to a service outage or other service line 
call. This will make it possible for the personnel to safely initiate 
correct repair procedures. The new marking requirement is expected to 
reduce deaths and injuries to gas distribution system personnel, and to 
reduce damage to the system and nearby property.
    The requirement to mark or otherwise identify services with EFVs is 
consistent with current industry practice. As a consequence, the 
benefits, costs, and net benefits are all expected to be $0 per EFV per 
year.

Performance Requirements

    The final rule sets performance requirements for all newly 
installed EFVs on single-family residential services operating at 10 
psig or greater. These performance requirements are to be ensured 
through design, manufacturing, and testing by EFV manufacturers in 
accordance with an industry specification or with the manufacturer's 
written specifications.
    The performance requirements will help ensure the reliability of 
EFVs. Greater reliability will result in (1) the replacement of fewer 
EFVs by gas distribution systems and (2) an increase in the number of 
EFV actuations when there are catastrophic service line breaks. The 
primary benefit of the new performance requirements will be an 
increased average reliability of the EFVs on the market. This assumes 
that all EFVs currently on the market are not fully consistent with the 
new requirements, which appears to be the case. A secondary benefit 
will be the assurance that the quality of EFVs will not degrade (with 
respect to the performance characteristics covered by the new 
performance requirements) in the future.
    The new performance requirements for EFVs cover (1) rated maximum 
operating pressure, (2) the impact of external temperature, (3) sizing, 
(4) reduction in gas flow upon closure, and (5) inappropriate closure. 
The requirements for rated maximum operating pressure, the impact of 
external temperature, and sizing appear to be consistent with current 
industry practice. The benefits of the new performance requirements are 
expected to be between $15,675 and $1,254 per year. The costs are 
expected to be $0 per year. Consequently, the net benefits are expected 
to be between $15,675 and $1,254 per year.
    The net benefits calculated for the performance requirements do not 
include (1) the costs related to the redesign of EFVs, (2) the full 
monetary value of the benefits accruing to gas distribution companies 
that currently install EFVs, and (3) the monetary value of the benefits 
that will accrue to gas distribution companies that install EFVs in the 
future.

Present Value of the Net Benefits

    The net benefits for the new performance requirements are the sum 
of the net benefits of (1) EFV installation standards, (2) the marking 
requirements, and (3) the EFV performance requirements. Since the net 
benefits for the EFV installation standards and for the marking 
requirements are expected to be greater than $0 per year, while the net 
benefits for the new performance requirements are expected to be 
between $15,674 and $1,254 per year, the total net benefits for the EFV 
requirements specified in the final rule will be, at most, greater than 
$15,674, and, at least, greater than $1,254 per year. Discounted over 
50 years (the life of an EFV assumed by OPS) using a 7 percent discount 
rate, the present value of the total net benefits is expected to be, at 
most, greater than $223,768, and, at least, greater than $17,901. Since 
costs are $0, their present value is also $0 and the cost-to-benefit 
ratio is 0 at both the upper and lower bounds of the benefits.

Conclusion

    The positive present value of the expected net benefits, as well as 
the cost-to-benefit ratio of 0 at both the upper and lower bounds on 
the benefits, indicate that the performance standards presented in the 
final rule will be cost beneficial.

Regulatory Flexibility Act

    Based on costing assumptions discussed in the Cost/Benefit 
Analysis, this rule will not have an undue impact on small operators. 
Therefore, I certify under section 605 of the Regulatory Flexibility 
Act that the action will not have a significant economic impact on a 
substantial number of small entities.

E.O. 12612

    This rulemaking action will not have substantial direct effects on 
states, on the relationship between the federal government and the 
states, or on the distribution of power and responsibilities among the 
various levels of government. Therefore, in accordance with E.O. 12612 
(52 FR 41685; October 30, 1987), RSPA has determined that this final 
rule does not have sufficient federalism implications to warrant 
preparation of a Federalism Assessment.

National Environmental Policy Act

    RSPA has analyzed this action for purposes of the National 
Environmental Policy Act (42 U.S.C. 4321 et seq.) and has determined 
that this action would not significantly affect the quality of the 
human environment. An Environmental Assessment and a Finding of No 
Significant Impact are in the docket.

List of Subjects in 49 CFR Part 192

    Pipeline safety, Reporting and recordkeeping requirements.
    In consideration of the foregoing, Part 192 is amended as follows:

PART 192--[AMENDED]

    1. The authority citation for Part 192 continues to read as 
follows:


[[Page 31459]]


    Authority: 49 U.S.C. 5103, 60102, 60104, 60108, 60109, 60110, 
60113 and 60118; 49 CFR 1.53.
* * * * *
    2. Part 192 is amended by adding Sec. 192.381 to subpart H to read 
as follows:


Sec. 192.381  Service lines: Excess flow valve performance standards.

    (a) Excess flow valves to be used on single residence service lines 
that operate continuously throughout the year at a pressure not less 
than 10 psig must be manufactured and tested by the manufacturer 
according to an industry specification, or the manufacturer's written 
specification, to ensure that each valve will:
    (1) Function properly up to the maximum operating pressure at which 
the valve is rated;
    (2) Function properly at all temperatures reasonably expected in 
the operating environment of the service line;
    (3) At 10 psig:
    (i) Be sized to close at, or not more than 50 percent above the 
rated closure flow rate specified by the manufacturer; and
    (ii) Upon closure, reduce gas flow--
    (A) For an excess flow valve designed to allow pressure to equalize 
across the valve, to no more than 5 percent of the manufacturer's 
specified closure flow rate, up to a maximum of 20 cubic feet per hour; 
or
    (B) For an excess flow valve designed to prevent equalization of 
pressure across the valve, to no more than 0.4 cubic feet per hour; and
    (4) Not close when the pressure is less than the manufacturer's 
minimum specified operating pressure and the flow rate is below the 
manufacturer's minimum specified closure flow rate.
    (b) An excess flow valve must meet the applicable requirements of 
Subparts B and D of this part.
    (c) An operator must mark or otherwise identify the presence of an 
excess flow valve in the service line.
    (d) An operator should locate an excess flow valve beyond the hard 
surface and as near as practical to the fitting connecting the service 
line to its source of gas supply.
    (e) An operator should not install an excess flow valve on a 
service line where the operator has prior experience with contaminants 
in the gas stream, where these contaminants could be expected to cause 
the excess flow valve to malfunction or where the excess flow valve 
would interfere with necessary operation and maintenance activities on 
the service, such as blowing liquids from the line.

    Issued in Washington, DC, on June 14, 1996.
D.K. Sharma,
Administrator, Research and Special Programs Administration.
[FR Doc. 96-15564 Filed 6-19-96; 8:45 am]
BILLING CODE 4910-60-P