Federal Motor Vehicle Safety Standards; Compressed Natural Gas Fuel Container Integrity

DEPARTMENT OF TRANSPORTATION

National Highway Traffic Safety Administration

49 CFR Part 571

[Docket No. 93-02; Notice 05]

RIN [2127-AF14]

Federal Motor Vehicle Safety Standards; Compressed Natural Gas

Fuel Container Integrity

AGENCY: National Highway Traffic Safety Administration (NHTSA),

Department of Transportation (DOT).

ACTION: Final rule.

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

SUMMARY: This rule establishes a new Federal motor vehicle safety

standard, Standard No. 304, Compressed Natural Gas Fuel Containers,

that specifies performance requirements applicable to compressed

natural gas (CNG) fuel containers: a pressure cycling test evaluates a

container's durability; a burst test evaluates a container's initial

strength; and a bonfire test evaluates a container's pressure relief

characteristics. In addition, the final rule specifies labeling

requirements for CNG containers. The purpose of this new standard is to

reduce deaths and injuries occurring from fires that result from fuel

leakage from CNG containers.

DATES: Effective Date: The Standard becomes effective March 27, 1995.

Incorporation by reference: The incorporation by reference of

certain publications listed in the regulations is approved by the

Director of the Federal Register as of March 27, 1995.

Petitions for Reconsideration: Any petition for reconsideration of

this rule must be received by NHTSA no later than October 26, 1994.

ADDRESSES: Petitions for reconsideration of this rule should refer to

Docket 93-02; Notice 5 and should be submitted to: Administrator,

National Highway Traffic Safety Administration, 400 Seventh Street SW.,

Washington, DC 20590.

FOR FURTHER INFORMATION CONTACT: Mr. Gary R. Woodford, NRM-01.01,

Special Projects Staff, National Highway Traffic Safety Administration,

400 Seventh Street, SW., Washington, DC 20590 (202-366-4931).

SUPPLEMENTARY INFORMATION:

Outline

I. Background

A. General Information

B. Previous Agency Rulemakings

II. Comments on the Proposal

III. Agency's Decision

A. Overview

B. Adopting Industry Standards

C. Pressure Cycling Test

D. Burst Test

1. Safety Factor

2. Hold Time Interval

3. Sequential Testing

4. Failure Criteria

E. Bonfire Test

1. Performance Requirements

2. Types of Pressure Relief Devices

3. Shielding

4. Test Gas and Pressure

5. Wind Velocity and Direction

6. Bonfire Fuel

7. Bonfire Test Fuel Pan Depth

F. Labeling Requirements

G. Leadtime

H. Benefits

I. Costs

VI. Rulemaking Analyses

A. Executive Order 12866 and DOT Regulatory Policies and

Procedures

B. Regulatory Flexibility Act

C. Executive Order 12612 (Federalism)

D. National Environmental Policy Act

E. Civil Justice Reform

I. Background

A. General Information

Natural gas is a vapor that is lighter than air at standard

temperature and pressure.\1\ When used as a motor fuel, natural gas is

typically stored on-board a vehicle in cylindrical containers at a

pressure of approximately 20,684 kPa pressure (3,000 psi). Natural gas

is kept in this compressed state to increase the amount that can be

stored on-board the vehicle. This in turn serves to increase the

vehicle's driving range. Since natural gas is a flammable fuel and is

stored under high pressure, natural gas containers pose a potential

risk to motor vehicle safety.

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

\1\Standard temperature is 0 deg. Celsius or 32 deg. Fahrenheit

and standard pressure is 101.4 kiloPascals (kPa) or 147.7 pounds per

square inch (psi).

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

Vehicles powered by CNG have not been numerous to date, although

they are increasing. The number of CNG vehicles in the United States

more than doubled from 10,300 in 1990 to 23,800 at the end of 1992. The

number of CNG vehicles is projected to again double to an estimated

50,800 vehicles in 1994. As discussed in detail in a final rule

regarding CNG vehicles published on April 25, 1994, recent Federal

legislation, as well as the need to meet environmental and energy

security goals, will lead to greater increases in the production and

use of these vehicles. (59 FR 19648).

B. Previous Agency Rulemakings

On October 12, 1990, NHTSA published an advance notice of proposed

rulemaking (ANPRM) to explore whether the agency should issue Federal

motor vehicle safety standards (FMVSSs) applicable to CNG fuel

containers and the fuel systems of motor vehicles using CNG or

liquified petroleum gas (LPG) as a motor fuel. (55 FR 41561). The ANPRM

sought comment about the crash integrity of vehicle fuel systems, the

integrity of fuel storage containers, and pressure relief for such

containers.

On January 21, 1993, NHTSA published a notice of proposed

rulemaking (NPRM) in which the agency proposed to establish a new FMVSS

specifying performance requirements for vehicles fueled by CNG. (58 FR

5323). The proposal was based on comments received in response to the

ANPRM and other available information. The NPRM was divided into two

segments: (1) vehicle requirements that focus on the integrity of the

entire fuel system, and (2) equipment requirements that focus on the

fuel containers alone.

NHTSA decided to model the proposed requirements applicable to CNG

fueled motor vehicles on Standard No. 301, Fuel System Integrity.

Standard No. 301 specifies performance requirements for vehicles that

use fuel with a boiling point above 32 deg.Fahrenheit (i.e., fuels

that are liquid under standard temperature and pressure). Vehicles

manufactured to use only CNG are not subject to Standard No. 301 since

CNG has a boiling point below 32 deg.F. Standard No. 301 limits the

amount of fuel spillage from ``light vehicles''\2\ during and after

frontal, rear, and lateral barrier crash tests and a static rollover

test. The Standard also limits fuel spillage from school buses with a

GVWR over 10,000 pounds after being impacted by a moving contoured

barrier at any point and any angle. By basing the CNG rulemaking on

Standard No. 301, the agency believed that passengers of CNG vehicles

would be afforded a level of safety comparable to that provided

passengers of vehicles fueled by gasoline or diesel fuel.

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

\2\Light vehicles include passenger cars, multipurpose passenger

vehicles (MPV's), trucks, and buses with a gross vehicle weight

rating (GVWR) of 10,000 pounds or less.

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

With respect to the ``vehicle'' requirements for CNG vehicles,

NHTSA proposed that the fuel system integrity requirements would

include frontal, rear, and lateral barrier crash tests for light

vehicles, and a moving contoured barrier crash test for large school

buses. The agency proposed that fuel system integrity would be

determined by measuring the fuel system's pressure drop after the crash

test rather than fuel spillage, since CNG is a vapor and not a liquid.

The allowable pressure drop for CNG fueled vehicles would be

equivalent, as measured by the energy content of the lost fuel, to the

allowable spillage of gasoline during Standard No. 301 compliance

testing.

With respect to the ``equipment'' requirements for CNG containers,

NHTSA proposed a definition for ``CNG fuel tank'' and performance

requirements that would apply to all such fuel containers manufactured

for use as part of a fuel system on any motor vehicle, including

aftermarket containers.\3\ Thus, while vehicles with a GVWR over 10,000

pounds (other than school buses) would not be subject to Standard No.

303, the CNG containers in those vehicles would be subject to the

equipment requirements. The agency proposed that each CNG container

would be subject to a pressure cycling test to evaluate container

durability and a pressure burst test to evaluate the container's

initial strength as well as its resistance to degradation over time. In

addition, the NPRM proposed requirements to regulate how the container

``vents'' its contents under specified conditions of elevated

temperature and pressure.

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

\3\Among the terms used to describe CNG fuel tanks are tanks,

containers, cylinders, and high pressure vessels. The agency will

refer to them as ``containers'' throughout this document.

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

II. Comments on the Proposal

NHTSA received a large number of comments to the docket addressing

the CNG proposal. The commenters included manufacturers of CNG

containers, vehicle manufacturers, trade associations, other CNG-

oriented businesses, research organizations, State and local

governments, the United States Department of Energy, and energy

companies. In addition, NHTSA met with the Compressed Gas Association

(CGA) and the Natural Gas Vehicle Coalition (NGVC) and had telephone

conversations and meetings with some of the commenters. A record of

each of these contacts may be reviewed in the public docket.

The commenters generally believed that a Federal safety standard

regulating the integrity of CNG fuel systems and fuel containers is

necessary and appropriate. In fact, some commenters, including the CGA,

the NGVC, and CNG container manufacturers stated that NHTSA should

issue a Federal standard as soon as possible to facilitate the safe and

expeditious introduction of CNG fueled vehicles. With respect to the

equipment requirements, the commenters generally believed that Federal

requirements about CNG fuel container integrity are needed and should

be implemented as quickly as possible. The CNG vehicle industry, led by

CGA and NGVC, expressed concern that lack of Federal regulations has

created a problem for the industry, given the issuance of potentially

conflicting industry and State regulations. Therefore, these commenters

stated that CNG container manufacturers may not know the appropriate

standards to which they should manufacture their containers. In

contrast, the American Automobile Manufacturers Association (AAMA)

stated that the vehicle system requirements are sufficient to regulate

the overall integrity of CNG fueled vehicles and that separate

requirements for CNG fuel containers are not needed. Nevertheless, AAMA

provided detailed comments about the container proposal in case the

agency decided to issue separate container requirements.

The commenters addressed a variety of issues discussed in the NPRM.

These issues include the appropriateness of adopting the American

National Standards Institute (ANSI) voluntary industry standard known

as NGV2;\4\ the pressure cycling requirements and test procedures; the

burst requirements and test procedures, including the proposed safety

factor, hold time interval, and need for sequential testing; the

pressure relief requirements and test procedures, including types of

pressure relief devices, shielding, test gas, test pressure, test fuel,

and fuel pan depth; labeling requirements; leadtime; costs; and

benefits.

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

\4\NGV2 is a recently issued voluntary industry standard that

was adopted by the ANSI and addresses CNG fuel containers. It was

developed by an industry working group that included container

manufacturers, CNG users, and utilities.

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

NHTSA issued an SNPRM proposing to pattern the burst requirements

more closely on NGV2, based on its consultation with other Federal

agencies, its review of comments to the January 1993 proposal, and

other available information. (58 FR 68846, December 29, 1993). NHTSA

proposed a burst test that would link the use of particular designs and

materials to compliance with safety factors tailored to those designs

and materials. NHTSA requested comment on the appropriateness of

requiring CNG containers to meet design and material requirements, such

as those specified in NGV2, and to meet safety factors tailored to

those requirements. As an alternative approach, the agency asked

whether it should specify a catch-all high end safety factor for any

container whose design and materials are not specified in NGV2.

Most commenters supported the proposal to incorporate NGV2 into the

Federal standard. However, AAMA and Ford opposed the design and

material specific approach of NGV2.

III. Agency's Decision

A. Overview

In today's final rule, NHTSA is issuing a new Federal motor vehicle

safety standard, Standard No. 304, Compressed Natural Gas Fuel

Containers, that specifies performance requirements applicable to a CNG

fuel container's durability, strength, and venting. A pressure cycling

test evaluates a container's durability by requiring a container to

withstand, without any leakage, 18,000 cycles of pressurization and

depressurization. This requirement helps to ensure that a CNG container

is capable of sustaining the cycling loads imposed on the container

during refuelings over its entire service life. A burst test evaluates

a container's initial strength and resistance to degradation over time.

This requirement helps to ensure that a container's design and material

are appropriately strong over the container's life. A bonfire test

evaluates a container's pressure relief characteristics when pressure

builds in a container, primarily due to temperature rise. In addition,

the final rule specifies labeling requirements for CNG fuel containers.

As previously mentioned, the agency has issued a final rule

establishing a new Federal motor vehicle safety standard, Standard No.

303, Fuel System Integrity of Compressed Natural Gas Vehicles, that

specifies vehicle performance requirements applicable to the fuel

system of a CNG fueled vehicle. As explained in that final rule, the

fuel system integrity requirements are comparable to those requirements

in Standard No. 301. Like that Standard, the new requirements limit the

amount of fuel leakage in specified frontal, rear, and lateral barrier

crash tests for light vehicles and a moving contoured barrier crash

test for school buses with a GVWR over 10,000 pounds.

NHTSA believes that CNG containers must be evaluated in all

possible failure modes and environments to which they may be subjected.

Since the requirements contained in today's final rule do not address

all these situations, the agency is currently investigating other

possible requirements for CNG fuel containers and anticipates issuing a

SNPRM that would propose performance requirements applicable to such

characteristics as a CNG fuel container's internal and external

resistance to corrosion, brittle fracture, fragmentation, and external

damage caused by incidental contact with road debris or mechanical

damage during the vehicle's operation. The agency tentatively believes

that these additional performance requirements are critical for

determining a CNG container's safety. In addition, the agency

anticipates proposing additional labeling requirements that should

provide critical safety information about inspecting a CNG container

and its service life.

NHTSA notes that it has no statutory authority to regulate certain

aspects involving CNG containers, including inspection requirements

during the manufacturing process, in-use inspection requirements, and

retest requirements during use.

B. Adopting Industry Standards

In the NPRM, NHTSA explained its decision to propose pressure

cycling and burst tests and requirements. While the agency's proposal

was based on NGV2, the agency decided not to propose certain provisions

of the voluntary industry standard that the agency tentatively believed

might unreasonably restrict future designs. Similarly, NHTSA decided

not to propose regulations issued by the Research and Special Programs

Administration (RSPA)\5\ for CNG storage containers used on motor

vehicles, explaining that the RSPA regulations do not address the

conditions unique to the motor vehicle environment (e.g., increased

cycling due to refueling and pressure relief when the cylinder is less

than full). NHTSA further explained that in contrast to RSPA, NHTSA

does not typically regulate design and materials since NHTSA is

statutorily directed to issue performance-based safety standards.

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

\5\RSPA is an administration within the United States Department

of Transportation that among other things regulates the

transportation of hazardous materials.

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

NGVC and several CNG container manufacturers stated that NHTSA

should adopt the voluntary industry standard that has been developed by

the CNG industry working group. In support of this request, the

American Gas Association (AGA) cited a 1982 Office of Management and

Budget Circular that states ``It is the policy of the Federal

Government to (a) Rely on voluntary standards * * * whenever feasible

and consistent with law and regulation pursuant to law * * *.'' AGA and

NGVC believed that the voluntary standards provide a higher level of

safety than the regulations proposed by NHTSA. They further stated that

if NHTSA were unable to adopt NGV2 due to its prescriptive nature, then

NHTSA should still allow automobile and equipment manufacturers the

option of certifying to the industry standard by referencing NGV2 in

the regulations.

In promulgating a CNG container standard, NHTSA has sought to the

extent possible to adopt the tests and requirements set forth in NGV2.

NHTSA was limited in its ability to do this by the National Traffic and

Motor Vehicle Safety Act (Safety Act, 49 U.S.C. 30111), which commands

the agency to issue ``motor vehicle safety standards'' as minimum

standards of motor vehicle performance that are practicable, meet the

need for motor vehicle safety, and are stated in objective terms. NHTSA

found it necessary to modify certain elements of NGV2 to be consistent

with this statutory mandate. For instance, the agency has not

incorporated those aspects of NGV2 that are stated in nonobjective

terms (e.g., a container shall not show ``evidence'' of deterioration

or failure) NHTSA has decided to incorporate NGV2's design and material

requirements since the agency has been unable to find or develop a

meaningful dynamic performance requirement that would adequately

evaluate a container's initial strength and susceptibility to

degradation over time. The agency believes that the requirements are no

more specific than necessary to achieve these safety purposes.

NHTSA notes that it would be impermissible under the Safety Act for

the agency to adopt FMVSS provisions referencing NGV2 in its entirety

and stating that automobile and equipment manufacturers had the option

of certifying compliance to NGV2 by referencing this voluntary industry

standard. The Safety Act provides for manufacturer self-certification

with respect to FMVSSs only. To be part of a FMVSS, the provisions of a

voluntary industry standard must fully meet all of the requirements of

the Safety Act. Since all of NGV2 does not meet these requirements,

NGV2 may not be incorporated in its entirety. Even if NGV2 met these

requirements, NGV2 could not be incorporated in the FMVSS except to the

extent that the FMVSS made compliance with NGV2 mandatory.

C. Pressure Cycling Test

In the NPRM, NHTSA proposed pressure cycling requirements that

would require that the fuel container withstand a cycling test at

ambient temperature, without any leakage or deformation exceeding one

percent of any circumference. In the test, the container would be

hydrostatically pressurized to the service pressure, then to not more

than 10 percent of the service pressure, for 13,000 cycles. The

container would next be hydrostatically pressurized to 125 percent of

the service pressure, then to not more than 10 percent of the service

pressure, for 5,000 cycles. The cycling rate would not exceed ten

cycles per minute.

1. Number of Cycles

The proposed cycling requirements were intended to establish

minimum levels of safety performance for the durability of CNG fuel

containers used in motor vehicles. The agency stated its tentative

belief that the requirements are consistent with provisions in NGV2 and

with RSPA regulations for containers used to transport CNG. The agency

believed that the pressure cycling requirement would help to assure

that a CNG container is capable of sustaining the cycling loads imposed

on the container during refuelings. The number of cycles specified in

the proposal, 13,000 plus 5,000, is representative of four refuelings

per day, 300 days per year, for 15 years.

AAMA, Norris, and Thomas commented on the number of pressure

cycles. These commenters stated that the proposed number of cycles was

excessive and not representative of the actual operating conditions the

CNG containers would typically experience. AAMA and Norris stated that

cycling the container at 125 percent of service pressure for 5,000

cycles would be adequate. Thomas made inconsistent statements about the

appropriate number of cycles. On the one hand, it stated that 9,000

cycles at service pressure would be more reasonable than the proposed

number of cycles. On the other hand, it stated that the agency should

adopt NGV2 which specifies 18,000 cycles.

After reviewing the comments and other available information, NHTSA

continues to believe that the proposed number of pressure cycles

accurately represents the extreme conditions that CNG fuel containers

could experience during their lifetime, with a margin of safety. This

is based on the large number of cycles to which fleet vehicles are

subjected. The agency believes that the 5,000 cycles suggested by AAMA

and Norris would not ensure the safety of vehicles that experience

multiple refuelings each day, such as taxis and other fleets. NHTSA

further notes that the number of cycles being adopted is consistent

with the cycles in NGV2 and therefore establishes a minimum level of

safety that is consistent with NGV2, a standard supported by a large

majority of the commenters. Accordingly, the agency has determined that

a CNG fuel container will be subject to 18,000 pressure cycles.

2. Failure Criteria

In the NPRM, NHTSA proposed that a CNG fuel container would have to

meet two test criteria to pass the pressure cycling test: (1) No

leakage, and (2) no permanent circumferential deformation greater than

one percent. The agency proposed these two criteria to provide

objective means of evaluating a container's durability during

compliance testing. NHTSA adopted the no leakage portion of the

proposal from NGV2's pressure cycling test. The one percent deformation

level, which is not in NGV2's pressure cycling test, was based on the

Society of Automotive Engineers (SAE) Recommended Practice J10, August

1985, a requirement involving the performance of metal air brake

reservoirs. The agency proposed a limit on circumferential deformation

to aid in determining when a container's failure was impending.

No commenters objected to the no leakage criterion. Accordingly,

the agency has adopted the no leakage requirement in the final rule.

The agency believes that specifying that containers ``shall not leak''

provides an objective measure that will ensure that a container

maintains its integrity by retaining its contents under pressure.

Sixteen commenters addressed the issue of the allowable

circumferential deformation criterion. The commenters were NGVC,

Brunswick, Pressed Steel Tank (PST), Structural Composites Industry

(SCI), Tecogen, CGA, AAMA, Amoco, Alusuisse, Oklahoma Gas, ARC,

Flxible, Fiber Dynamics, Norris, Comdyne, and EDO. All the commenters,

except Brunswick, believed that the agency should not include a

deformation requirement in the pressure cycling or burst tests. The

commenters believed that the test requirement is not appropriate for

all container materials and designs. They stated that due to the nature

of the different materials used in these containers, and their

different rates of deformation under load, some materials such as

fiberglass, would deform more than others, such as steel. The

commenters also stated that deformation was not an indicator of

impending failure and that the SAE brake reservoir test was not

appropriate for a CNG fuel container application.

NHTSA has decided not to adopt the one percent circumferential

deformation requirement. In proposing this criterion, NHTSA tentatively

concluded that it would be an appropriate indicator of the fuel

container's durability characteristics. However, as the comments note,

it is not an appropriate criterion because of the differing

construction and materials used for CNG fuel container applications.

After reviewing the comments and other available information, the

agency now believes that limiting the circumferential deformation is

not a meaningful way to determine a container's strength or impending

failure, since the larger deformation experienced by some materials

does not necessarily represent these characteristics. Instead, the

agency believes that the no-leakage requirement, by itself, is the

appropriate criterion to define a container failure, after being

subjected to the pressure cycling test.

Brunswick further commented that some container designs, such as

full-wrapped composite containers, would deform in the axial direction

in addition to the circumferential direction. To account for axial

deformation, Brunswick recommended allowing a maximum five percent

volumetric expansion of the container.\6\ Brunswick stated that this

test is used to assure that the container material exhibits elastic

behavior at expected operating conditions.

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

\6\Both RSPA's standards and NGV2 incorporate the concept of

volumetric expansion. In these standards, the volumetric expansion

is measured when hydrostatic testing is performed on the container

at 1.50 to 1.67 times the service pressure. This test is a non-

destructive one, i.e., the container may be put into service after

it is tested.

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

NHTSA agrees with Brunswick's statement that some container designs

deform in the axial direction. Nevertheless, the agency believes that

measuring volumetric expansion would not provide an appropriate measure

of a container's impending failure in a destructive test (i.e., where

the container cannot be used again). In addition, the NPRM provided no

notice to amend the standard to measure such expansion in the axial

direction. Since the pressure cycling and burst tests being adopted in

this rule are capable of evaluating a CNG container's durability, the

agency believes that another non-destructive test would be redundant

and therefore is not needed. The agency further notes that the five

percent maximum level of expansion would not provide a meaningful

measure of a container's impending failure, since this level is based

on a container's performance under less stringent test conditions.

D. Burst Test

1. Safety Factor

With respect to the burst test, NHTSA proposed that a CNG fuel

container would have to withstand an internal hydrostatic pressure of

3.50 times the service pressure for 60 seconds, without any leakage or

circumferential deformation over one percent. The multiple of the

internal hydrostatic pressure, 3.50, is known as the safety factor. The

agency tentatively concluded that the burst test, together with a

pressure cycling test, would be sufficient to assure adequate levels of

safety performance for both the strength and durability of CNG fuel

containers used in motor vehicles.

The proposal of a burst test with a safety factor was based in part

on NGV2. NGV2 specifies several sets of detailed material and design

requirements. For each set of those requirements, NGV2 specifies a

unique safety factor for calculating the internal hydrostatic pressure

that the container must withstand. The safety factors range from 2.25

to 3.50, depending on the material and design involved. To satisfy this

aspect of NGV2, a container must meet both the material and design

requirements as well as the burst test.

NGV2 specifies four types of container designs. A Type 1 container

is a metallic noncomposite container. A Type 2 container is a metallic

liner over which an overwrap such as carbon fiber or fiberglass is

applied in a hoop wrapped pattern over the liner's cylinder sidewall. A

Type 3 container is a metallic liner over which an overwrap such as

carbon fiber or fiberglass is applied in a full wrapped pattern over

the entire liner, including the domes. A Type 4 container is a non-

metallic liner over which an overwrap such as carbon fiber or

fiberglass is applied in a full wrapped pattern over the entire liner,

including the domes.

The agency did not propose adoption of the material and design

requirements of NGV2. Instead, the agency proposed a single safety

factor of 3.50 for all containers, regardless of their materials or

design. It tentatively concluded that the factor would not impede

technological development, yet would assure an acceptable level of

safety for all containers.

CNG container manufacturers, CNG trade associations (NGVC and AGA),

utility companies, the American Automobile Manufacturers Association

(AAMA) and other commenters addressed the issue of the safety factor.

Most commenters disagreed with the agency's proposal to require that

all containers meet the same safety factor.

NGVC, AGA, and the CNG container manufacturers generally believed

that the material and design of the fuel container need to be taken

into account in establishing an appropriate safety factor, if safe,

cost-effective, and light-weight containers are to be produced.

Establishing an overly high factor for a given combination of material

and design could result in unnecessarily over-designed, heavy

containers, according to these commenters. They believed that some

materials, such as fiberglass, need a higher safety factor because they

degrade faster over time. In contrast, a material such as steel

maintains its strength for a longer time, and therefore containers made

of it could be made safely with a lower safety factor.

Many of these commenters recommended that NHTSA adopt the safety

factors specified in NGV2. They stated that compared to the regulations

proposed by NHTSA, the NGVC voluntary industry standard provides a more

appropriate level of safety, given the need to specify safety factors

based on the design and materials used.

However, several commenters disagreed with certain safety factors

specified in NGV2. CGA, PST, SCI, and NGV Systems supported a higher

safety factor for containers using unproven materials. In particular,

they were concerned with containers reinforced with carbon fiber

overwrap, for which NGV2 specifies a 2.25 safety factor for all carbon

reinforced containers, Types 2, 3, and 4.

NGV Systems stated that a safety factor of 2.25 constitutes an

``unacceptable safety risk,'' given the industry's limited experience

with carbon fiber and lack of a significant data base demonstrating

this materials safety and reliability. Accordingly, NGV Systems

supported a safety factor of 3.5 for what it termed unproven designs,

which may then be lowered as more experience and data accumulate. CGA

recommended safety factors of 2.5 for all Type 2 containers and 3.33

for all Type 3 and 4 containers, stating that these are used on all

fiber reinforced compressed gas containers now in commercial use. CGA

indicated that unlike other fiber overwrap used in the past for

transportation pressure vessels, there is no commercial experience with

the safety of carbon fiber reinforced containers for motor vehicle

applications to justify a 2.25 safety factor for such containers. CGA

stated that NGV2 does not adequately address damage tolerance concerns

for carbon reinforced fully wrapped containers with low safety factors.

PST recommended 3.33 for carbon fiber Types 3 and 4 containers. That

commenter recommended such conservative safety factors until

substantial data are accumulated on the use of carbon fiber containers

in actual service. SCI provided similar comments, and recommended

safety factors of 3.33 for the fully wrapped containers, which are

Types 3 and 4.

Three commenters stated that a single safety factor was

appropriate. CNG Pittsburgh, a consulting firm, stated that a safety

factor of 3.50 is conservative but reasonable for CNG fuel containers.

AAMA stated that adopting NGV2's approach with various safety factors

depending on the material and design involved would limit a

manufacturer's choice of container designs and materials. EDO

recommended a safety factor of 2.5 for all containers.

NHTSA decided to issue an SNPRM proposing to pattern the burst

requirement more closely on NGV2, based on its consultation with other

Federal agencies, its review of comments to the January 1993 proposal,

and other available information. In explaining its reason for issuing

the SNPRM, NHTSA stated that there did not appear to be any procedures

that could adequately test a container's susceptibility to degradation

over time. Therefore, it believed that specifying a single safety

factor would not protect in all instances against these problems since

the strength of some containers is dependent on the specific material

and method of design. Therefore, NHTSA decided to propose a burst test

that would link the use of particular designs and materials to

compliance with safety factors tailored to those designs and materials.

The agency tentatively concluded that such an approach might be

necessary to ensure the safe performance of pressure vessels used for

fuel containers. The agency further noted that international standards

addressing CNG fuel containers, including regulations of Transport

Canada and those being drafted by the International Standards

Organization (ISO) link the use of particular designs and materials

with strength requirements suitable for those designs and materials.

In the SNPRM, NHTSA requested comment on the appropriateness of

requiring CNG containers to meet design and material requirements, such

as those specified in NGV2, and to meet safety factors tailored to

those requirements. The agency also asked about the effect of adopting

NGV2 on future container technology, since the only way a container

manufacturer could comply with the Federal standard would be by

producing a container that uses those materials and designs specified

in NGV2 if the agency incorporated NGV2's material and design

provisions in the FMVSS. As an alternative approach, the agency asked

whether it should specify a catch-all high end safety factor for any

container whose design and materials are not specified in NGV2.

NHTSA received 18 comments to the December 1993 SNPRM about

adopting the design and material specific approach of NGV2. Sixteen

commenters, including NGVC/AGA, CGA, CNG container manufacturers,

public utilities, and two bus manufacturers supported the proposal to

incorporate NGV2 into the Federal standard. Eleven commenters supported

the safety factors in NGV2. Five others were concerned about the level

of some safety factors in NGV2 or the use of relatively new materials,

such as carbon fiber. CGA and SCI referenced their earlier comments to

the NPRM, again recommending safety factors of 2.5 for all Type 2

containers and 3.33 for all Type 3 and Type 4 containers. AAMA and Ford

opposed the design and material specific approach of NGV2. AAMA stated

that some of NGV2's requirements limit opportunities for future

development of advance container design or materials that may not fit

in the specifications in NGV2. No commenter favored having a catch-all

high end safety factor.

Based on the available information, NHTSA has decided to require

CNG containers to meet the safety factors applicable to the design and

material requirements specified in NGV2, except for carbon fiber.

Specifically, the agency is specifying separate safety factors for

containers using various materials (e.g., fiberglass, carbon, steel,

aluminum) and different designs (non-composite, hoop wrapped or full

wrapped composite containers, and welded). The agency believes that

this approach will result in the manufacture of safe containers for CNG

powered vehicles.

NHTSA has decided to adopt the specific safety factors and related

requirements set forth in NGV2, except for those safety factors

specified for carbon fiber. While NGV2 currently specifies a safety

factor of 2.25 for Type 2, 3, and 4 carbon fiber containers, NHTSA has

decided to specify a safety factor of 2.5 for Type 2 carbon fiber

containers and 3.33 for the Type 3 and 4 carbon fiber containers. The

agency is requiring a higher safety factor for Type 3 and 4 containers

since the fibers on those containers carry a greater proportion of the

load than on Type 2 containers.

NHTSA made this decision after reviewing all of the comments and

information obtained in response to both the NPRM and SNPRM; meetings

with container manufacturers, CGA and NGVC/AGA; and discussions with

other Federal agencies, including RSPA. Comments and information were

presented to support safety factors for carbon fiber containers,

ranging from 2.25 to 3.5. Brunswick, in particular, submitted

substantial test data and other technical information in support of

NGV2's 2.25 safety factor for carbon fiber, including testing it

performed on such containers which showed favorable results. RSPA

recommended a safety factor of not less than 3.0 for carbon fiber,

which is consistent with its FRP-1 and FRP-2 standards.

Notwithstanding comments supporting the 2.25 safety factor, NHTSA

has determined that under its statutory mandate, it is necessary to

specify higher safety factors for carbon fiber containers. In adopting

these more stringent requirements, NHTSA sought the advice of RSPA,

which has accumulated significant experience and expertise through its

efforts to regulate the safety of pressure vessels used to transport

hazardous materials. Specifically, NHTSA has adopted RSPA's

recommendation not to specify the 2.25 safety factor for carbon

composite containers.

The more stringent safety factors being adopted are consistent with

RSPA's longstanding approach to initially adopt conservative

requirements and subsequently modify the requirements, if further real-

world safety data become available supporting less stringent

regulations. NHTSA has determined that applying this approach to the

safety factors for carbon fiber containers is necessary, since carbon

fiber containers have not been used extensively in motor vehicle

applications. The agency believes that the higher safety factors are

justified until further data are developed and become available on the

use of carbon fiber containers in motor vehicle applications.

NHTSA acknowledges that using such a safety-oriented approach may

result in costlier and heavier carbon fiber containers. However, the

agency believes that the requirements being adopted will not preclude

the introduction and effective use of this new technology. Overall, the

agency believes that the safety factors being specified for carbon

fiber containers, along with the remaining safety factors it has

adopted from NGV2 for other materials, will result in safe CNG

containers.

As for AAMA's comment, NHTSA shares that association's concerns

about restricting future developments. However, based on comments by

the container manufacturers and other Federal agencies, the agency

believes that few, if any, designs beyond those accounted for in NGV2

are planned. If a new container technology is developed, the agency

will evaluate its safety in the context of a petition for rulemaking to

amend the Federal safety standard.

NHTSA has decided not to adopt the catch-all high level safety

factor, which could allow containers incorporating materials or designs

that have not been incorporated in NGV2 and thus might be detrimental

to safety. The agency further believes that it would be inappropriate,

at this time, to add a catchall factor. While such a proviso would

facilitate innovation and design change, the agency agrees with

commenters that specifying such a catchall might be detrimental to

safety, since untested designs and materials would be permitted.

2. Hold Time Interval

In the NPRM, NHTSA proposed that during the burst test, elevated

pressure would have to be sustained for 60 seconds. The agency noted

that while RSPA regulations also specify a 60-second period, NGV2

requires a 10-second hold time interval once the maximum pressure is

obtained. The agency believed that because NGV2 includes additional

tests to qualify container designs and the agency was not proposing

these additional tests, a shorter hold time would not be suitable.

NHTSA received six comments addressing the appropriate hold time

interval. All commenters except EDO believed the 60 second hold time

requirement was not necessary. EDO stated that the requirement was

``tough but reasonable.'' NGVC, Brunswick, PST, and ARC stated that

specifying the hold time at 60 seconds instead of 10 seconds would not

compensate for the lack of other NGV2 required tests. NGVC stated that

the ten second hold time interval is not intended as a test of

container strength, but as the time for the pressure in the container

to stabilize. PST stated that along with the 3.5 safety factor, the 60

second hold time would make an already conservative test even more

stringent.

After reviewing the comments and other available information, NHTSA

has decided to specify a hold time of 10 seconds instead of 60 seconds.

The agency notes that the proposal was based on a misperception of the

hold time requirement's purpose. As the commenters stated, the hold

period is included only to stabilize the pressure. It is not used as a

surrogate for initial burst strength. Therefore, the reduction in hold

time will not affect the test's stringency. In addition, the agency

anticipates issuing a SNPRM that would propose additional performance

requirements to evaluate other aspects of a CNG fuel container's

integrity.

3. Sequential Testing

In the NPRM, NHTSA proposed that a container that passed the

pressure cycling test would then be subjected to the burst test. In

proposing that the same fuel container be used in both the pressure

cycling and burst tests, the agency believed that it would be

appropriate to establish that the fuel container maintained its initial

strength after being subject to the durability test.

Seven commenters addressed the issue of using the same container

for both the pressure cycling and burst tests. NGVC, AAMA, Comdyne,

Pressed Steel Tanks, and Amoco stated that requiring the same fuel

container for both tests would be unrealistic and overly stringent,

because in real world situations, a container would not be subject to

pressure cycling and burst conditions sequentially. They stated that

otherwise unnecessary material would have to be added to strengthen the

container so it could meet the burst test requirement after the

pressure cycling test. These commenters believed such additional

material would significantly increase the container's cost and weight

to the extent that the container would no longer be economically viable

to produce. They further stated that most containers that are currently

produced to meet NGV2 or RSPA requirements would not be able to meet

this requirement. In contrast, EDO and Metropolitan Suburban Bus

Authority (MSBA) favored the use of sequential testing.

After reviewing the comments and other available information, NHTSA

has decided not to require sequential testing. The agency believes that

using different containers in the pressure cycling and burst tests will

provide an adequate measure of both the container's initial strength

and its durability over its life, without imposing new cost burdens on

the industry. The agency notes that such testing is consistent with the

way in which industry currently tests under both NGV2 and RSPA

standards. The agency further notes that in testing for compliance with

some FMVSSs, the agency allows a manufacturer to use a separate vehicle

or component for different tests within a standard. For example, three

vehicles are crashed in Standard No. 301, and different brake hoses are

used for various tests in Standard No. 106, Brake Hoses.

4. Failure Criteria

In the NPRM, NHTSA proposed that to pass the burst test, a

container would have to meet the same two performance criteria as in

the pressure cycling test: (1) No leakage, and (2) no permanent

circumferential deformation of more than one percent. The purpose of

these requirements was to provide objective means to evaluate a

container's compliance strength. NGV2 includes the no leakage

criterion, but not the one percent circumferential deformation

criterion. As explained in the section on the pressure cycling test,

the deformation requirement was based on SAE Recommended Practice J10,

August 1985, which addresses the performance of metal air brake

reservoirs. The agency proposed a circumferential deformation limit to

aid in determining a container's impending failure.

After reviewing the comments, NHTSA is adopting the no leakage

criterion to evaluate failure of the burst test. The agency has decided

not to adopt the one percent deformation criterion because the agency

believes that circumferential deformation is not a meaningful measure

of a fuel container's impending failure in the burst test. See the

section above regarding the pressure cycling test for a more

comprehensive discussion about the agency's decision not to adopt the

pressure deformation criterion.

E. Bonfire Test

1. Performance Requirements

In the NPRM, NHTSA proposed performance requirements for CNG fuel

containers to address the need to withstand high temperatures and

pressures without catastrophic failure. Large pressure increases due to

exposure to flames could cause the CNG to escape catastrophically and

result in an explosive fire. The agency proposed that the ability to

withstand high temperatures and pressures be provided by a pressure

relief device. More specifically, it proposed that compliance would be

determined by first pressurizing the fuel container to 100 percent of

service pressure with nitrogen or air and placing it over a bonfire

until the container's contents are completely vented through a pressure

relief device. A pressure relief device can prevent a container from

experiencing high pressure for long periods of time. The agency

proposed a second test to be conducted in the same manner, except the

container would be pressurized to 25 percent of the service pressure.

The second test would evaluate container performance when containers

are partially filled. The purpose of the test is to reduce the

explosion potential of CNG containers when exposed to high temperatures

and pressures.

The proposed requirements were based on NGV2. However, there were

two differences between the agency's proposal and NGV2. First, under

the NPRM, the container would be pressurized with nitrogen or air; in

NGV2, it is pressurized by CNG. Second, under the NPRM, all fuel

containers would be required to use a pressure relief device to

completely vent the container's contents; in NGV2, the test is run for

20 minutes or until the container is completely vented, whichever comes

first. Therefore, under NGV2, a manufacturer could establish compliance

either by a container successfully withstanding the test conditions for

20 minutes without bursting or by completely venting its contents by

means of a pressure relief device at some point during that 20 minute

period. In the NPRM, the agency sought comment about whether to allow

an alternative way of demonstrating compliance with the bonfire test

that did not depend upon a pressure relief device. Under the

alternative, a container would be considered to have passed the test if

it did not burst during the test period. Compliance with the

alternative would be achieved by designing a container so that it has

sufficient strength to enable it to sustain the heat and pressure

buildup during the test.

Eleven commenters addressed the issue of whether containers should

be required to have a pressure relief device. NGVC, EDO, ARC, Flxible,

Manchester, Thomas, and MSBA agreed with the proposal to require

containers to be equipped with such a device. They stated that a

pressure relief device is an integral part of a CNG container and that

its importance warrants a requirement that each container have one. In

contrast, Brunswick, Comdyne, Pressure Technology, and AAMA stated that

containers should not be required to have a pressure relief device

because such a requirement would be design restrictive. Brunswick and

Pressure Technology stated that the container should be required to

``safely vent'' its contents without rupturing, whether the venting is

done through a pressure relief device or the container wall. AAMA

stated that a container should pass the requirement if it possesses

enough strength to retain its contents throughout the test. ARC

believed that the container sidewalls should not be permitted to

rupture during the bonfire test.

After reviewing the comments, NHTSA has determined that each CNG

container must be equipped with a pressure relief device. This is

necessary because each CNG fuel container needs to possess a means of

releasing its contents in case the internal pressure or temperature

reaches a dangerous level. By requiring containers to be equipped with

a pressure relief device, the agency will ensure the safety of

individuals, such as vehicle occupants and rescue personnel, who would

be near a CNG vehicle in a fire. The agency notes that the conditions

experienced in the bonfire test may be less severe than certain real-

world crash situations. Therefore, the agency is adopting a more

conservative approach and requiring a pressure relief device for all

containers. In addition, such a requirement is consistent with the

practice of most container manufacturers and NGV2 which requires such a

device on all containers.

Based on the comments, NHTSA has decided to adopt NGV2's test

criteria that allows the test to be completed after 20 minutes or when

the container has completely vented, whichever comes first. Adopting

these criteria alters the test in that while still requiring a pressure

relief device, a container could comply with the bonfire test if it

either completely vents its contents by means of a pressure relief

device at some point during that 20 minute period or by successfully

retaining the container's entire contents without bursting for the

duration of the bonfire test (i.e., 20 minutes). The agency believes

that each criterion appropriately measures a container's ability to

withstand high temperature and pressure because the bonfire test

represents extreme conditions. The agency emphasizes that in either

case the CNG container must be equipped with a pressure relief device.

NHTSA disagrees with the approach advocated by AAMA, Brunswick and

Pressure Technology to allow containers to ``safely vent'' their

contents from an area other than the pressure relief device such as the

sidewall. The agency acknowledges that, as an alternative to a pressure

relief device, pressure relief can be accomplished by allowing the

overpressurized container to vent its contents at a controlled rate,

without fragmentation, through the container's sidewall. However, there

would be significant problems with this approach. First, it would not

afford as high a degree of safety as requiring a pressure relief

device. The agency continues to believe that the safest way to release

CNG from an overpressurized container is through a pressure relief

device because some sidewall ruptures could result in fragments being

propelled from the container. Second, it would raise potential

enforceability problems since the concepts of ``release its contents at

a controlled rate'' and ``rupture without fragmentation'' are difficult

to define objectively. Based on the above considerations, NHTSA has

decided to require each CNG fuel container to either completely vent

its contents through a pressure relief device or not burst when tested

in accordance with the test conditions.

2. Types of Pressure Relief Devices

The proposal did not specify the use of a particular type of

pressure relief device. The agency is aware of three types of devices

currently being used: (1) The rupture disc, which is designed to

release CNG in the container when it reaches a specific pressure, (2)

the fusible plug, which is designed to release CNG in the container

when it reaches a specific temperature, and (3) a device that combines

these two devices.

Four commenters recommended the use of specific types of pressure

relief devices. EDO recommended that the agency require the fusible

plug device and prohibit the rupture disc device. EDO stated that a

combination of hot conditions and overfill at the refueling pump could

cause a rupture disc to activate, releasing CNG and causing a

potentially dangerous situation. It further believed that the safety

factor in the burst test would be sufficient to prevent over

pressurization and that the pressure relief device should only open in

a fire situation. Flxible stated that the agency should require a

fusible plug to ensure pressure relief of partially filled containers

subject to heat or fire. NYCFD stated that the agency should prohibit

the combination fusible plug and rupture disc devices, claiming that

over-charged containers exposed to high ambient temperature are likely

to fail whether or not they are exposed to fire. Thomas commented that

the agency should require the combination fusible plug and rupture disc

device because it is required by NFPA 52.\7\

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

\7\NFPA 52, Standard for Compressed Natural Gas (CNG) Vehicular

Fuel Systems, is a voluntary standard adopted by the National Fire

Protection Association that specifies guidelines for the ``design

and installation of CNG engine fuel systems on vehicles of all types

including aftermarket and OEMs and to their associated fueling

(dispensing) systems.'' (NFPA 52, Sec. 1-1)

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

After reviewing the comments, NHTSA has concluded that the standard

should not specify the type of pressure relief device with which a

container may be equipped. The NPRM and SNPRM did not provide

sufficient notice for the agency to adopt such a specification as part

of this final rule. Further, the agency believes that the bonfire test,

which is performed at both 100 percent of service pressure and 25

percent of service pressure, will adequately evaluate a container's

ability to vent its contents in a high temperature/pressure situation.

In the first test, the combination of the 100 percent service pressure

condition and the high heat from the bonfire will cause the container's

pressure to increase rapidly. This test evaluates a container's ability

to vent its contents at high temperatures and pressures. In the second

test, the 25 percent service pressure condition and the heat will cause

the container's temperature to increase before the pressure in the

container reaches a critical point. This test evaluates a container's

ability to vent its contents at high temperatures, where the container

is at a less than full condition.

3. Shielding

NHTSA notes that there are two types of shielding that can affect

the performance of pressure relief devices in bonfire tests: (1)

``Vehicle-based protective shielding'' that is placed around the

container in the vehicle to protect the container from surrounding

heat, and (2) ``test shielding'' that is placed over the pressure

relief device to prevent flames from contacting the device. Test

shielding is, as the name suggests, installed only for the purpose of

conducting bonfire tests. Unlike vehicle-based protective shielding, it

is not used to affect real world performance.

In the NPRM, NHTSA recognized that some CNG vehicles may have

vehicle-based shielding installed to protect the containers from

exposure to heat. Nevertheless, the agency proposed that no vehicle-

based shielding be used during the bonfire test because Standard No.

304 is an equipment standard, and applies to CNG containers, not to

vehicles. Further, since the presence or amount of shielding could vary

from vehicle to vehicle, the agency tentatively concluded that the

containers should be tested in the worst case situation, i.e., without

any vehicle-based shielding. Nevertheless, the agency stated that it

did not want to discourage vehicle manufacturers from including

shielding in CNG vehicles as an added safety feature.

NHTSA received six comments addressing the use of vehicle-based

shielding during the bonfire test. PST, EDO, ARC, Ontario, and NGVC

agreed with the agency that vehicle-based shielding of the container

should not be used during the bonfire test. They believed that such

shielding could detract from or mask the results of the test. In

contrast, AAMA stated that ``[i]f a manufacturer chooses to add the

additional expense to protect the fuel tank from exposure to potential

flame, the protection ought to be allowed in any test as representative

of the tank's use in the vehicle.''

After reviewing the comments, NHTSA has decided not to permit

vehicle-based shielding of the container during the bonfire test. As

explained in the NPRM, the bonfire test is intended to evaluate the

container and not the vehicle. Since this is an equipment standard, the

tests are designed to ensure that the containers are safe for

installation in any vehicle, regardless of the amount of protective

vehicle shielding, if any, with which it is equipped. The agency

disagrees with AAMA's contention. Using vehicle shielding in compliance

testing would not ensure that a container could perform safely under

worst case conditions (i.e., no vehicle-based shielding of any type or

extent) that the container could encounter during its service life

(e.g., if the container is subsequently placed in a different vehicle).

Test shielding consists of a metal plate over the pressure relief

device and is permitted, but not required, under NGV2 for purposes of

the horizontal bonfire test. In the horizontal test, the CNG container

is positioned over the bonfire with its longitudinal axis in a

horizontal position. In the NGV2 vertical bonfire test (container

longitudinal axis in a vertical position), pressure relief device

shielding is also permitted, but not required, except where the CNG

container is fitted with a pressure relief device on both ends. In that

case, the bottom pressure relief device must be shielded. The goal is

to not allow flames to impinge directly on any relief device. This may

be done through test shielding, or by orienting the container so as to

avoid flame impingement on any pressure relief device. Without this

metal plate, the flames could contact the pressure relief device,

possibly causing it to vent the container prematurely. If this

occurred, the bonfire test results would neither evaluate the CNG

container as a whole nor accurately reflect the container's pressure

relief characteristics.

CGA and PST opposed allowing shielding of the pressure relief

device during the bonfire test. They commented that shielding the

pressure relief device during the bonfire test would not be

representative of a real-world crash fire situation. CGA stated that

allowing, but not requiring shielding to be placed around pressure

relief devices could produce non- repeatable results. PST stated that

excessive shielding around the pressure relief device could cause an

otherwise acceptable design to fail the test, but did not elaborate as

to how this could occur.

NHTSA has decided to require test shielding of the pressure relief

device during the horizontal bonfire test. The agency notes that the

purpose of this test is to replicate the effect of fires on the

pressure relief device and the fuel container as a system. Requiring

shielding will assure that the bonfire test is evaluating the fuel

container as a whole, rather than merely the pressure relief device,

since a flame that impinges on the pressure relief device, could

activate prematurely. Requiring shielding, rather than simply allowing

it, will assure repeatable and consistent test results. The rule also

requires shielding of the pressure relief device during the vertical

bonfire test, except where the container is fitted with a pressure

relief device on only one end. In that case, the container is

positioned with the pressure relief device on top, so as to avoid

direct contact with the flame.

4. Test Gas and Pressure

In the NPRM, NHTSA proposed that the CNG container be pressurized

with either nitrogen or air to 100 percent of service pressure for the

bonfire test. The agency acknowledged that NGV2 specifies the use of

CNG, but tentatively concluded that using nitrogen or air as the test

gas would be safer than using CNG.

AAMA and Tecogen recommended that CNG be used as the test gas.

Tecogen further commented that the container manufacturers have

historically conducted such tests using CNG and are therefore well

aware of the necessary safety precautions. It further stated that using

CNG as the test gas would reveal the pressure relief valve's

effectiveness with respect to the discharge rate. AAMA commented that

CNG should be used as the test gas because the thermal properties of

CNG differ from those of nitrogen and air and NGV2 specifies the use of

CNG as the test gas. AAMA also recommended that the CNG containers be

pressurized at the start of the test to 95 to 100 percent of service

pressure, but offered no rationale.

After reviewing the comments, NHTSA has determined that using CNG

as the test gas would better reflect the real-world conditions in a

fire, since the test gas would be the same as the gas used in CNG

containers. The agency notes that the bonfire test addresses the

responsiveness of the pressure relief device and that air and nitrogen

have different thermal properties than CNG. Therefore, the pressure

relief device might perform differently if air or nitrogen were used

instead of CNG. In the NPRM, the agency explained that using CNG as a

test gas might not be safe. These initial concerns have been allayed by

the comments indicating that manufacturers are aware of and accustomed

to taking the necessary safety precautions when using CNG as a test gas

to evaluate a container. NHTSA notes that it decided not to specify CNG

as the test gas in the CNG vehicle standard. Nevertheless, the agency

believes that differences in reaction to heat are important for the

bonfire test, which involves high temperatures, but not for crash

tests, which do not involve such temperatures.

NHTSA continues to believe that it is necessary to pressurize the

CNG container to 100 percent of service pressure at the outset of the

test. The agency has determined that the containers need to be tested

at full service pressure to represent the worst case scenario.

5. Wind Velocity and Direction

In the NPRM and SNPRM, NHTSA did not address the allowable wind

velocity and direction. The agency received comments from NGVC, CGA,

and PST stating that a limit should be placed on wind velocity to

increase the bonfire test's repeatability.

After reviewing the comments, NHTSA has decided to specify that the

average wind velocity at the container during the test may not exceed

2.24 meters per second (5 mph). The agency believes that permitting

higher crosswinds would vary or reduce the flame's heat. Therefore,

placing limits on the crosswind assures the test's repeatability and

the level of stringency that the agency anticipated in proposing this

test.

6. Bonfire Fuel

In the NPRM, NHTSA proposed that the fire for the bonfire tests be

generated using No. 2 diesel fuel. This fuel type was proposed so that

the standard would be consistent with the bonfire test in NGV2, which

also specifies this type of fuel.

NGVC, CGA, AAMA, and Norris commented that the agency should

specify a different fuel to generate the bonfire that is more

environmentally sound. CGA stated that the large amounts of smoke that

would be created by burning the diesel fuel are contrary to the

environmental objectives of developing CNG vehicles. NGVC and Norris

suggested using a CNG or propane grill for the test.

After reviewing the comments and other available information, NHTSA

has decided to specify the use of No. 2 diesel fuel in the final rule.

The agency is aware of the environmental problems associated with this

type of fuel and will further study whether other fuels should be used

to generate the bonfire test. However, until the agency can determine

that a different fuel is an appropriate replacement for diesel fuel,

the Standard will specify No. 2 diesel fuel for use in the bonfire

test.

7. Bonfire Test Fuel Pan Depth

In the NPRM, NHTSA proposed that the bonfire test pan containing

No. 2 diesel fuel be at least 100 centimeters (cm) deep. The agency

specified a depth to ensure that there would be an adequate amount of

fuel to run the test.

AAMA, Comdyne, CGA, Alusuisse, and PST commented that the fuel pan

depth was excessive. Alusuisse stated that a pan of the proposed size

would contain more than 1,000 liters of fuel. PST stated that a 100

millimeter (mm) depth would be more reasonable. CGA, AAMA, and Comdyne

stated that the depth of the fuel pan should not be specified so long

as a sufficient quantity of fuel is provided for the test.

The agency intended to propose a depth of 100 mm. However, due to a

typographical error, it proposed a depth of 100 cm. NHTSA agrees that a

fuel pan with a depth of at least 100 cm would be too deep. NHTSA also

agrees that the fuel pan's depth does not need to be specified,

provided that there is a sufficient amount of fuel to maintain the fire

for the duration of the test. Accordingly, the agency has removed the

requirement for fuel pan depth and has replaced it with the provision

that there be ``sufficient fuel to burn for at least 20 minutes.'' The

agency believes that this provision is consistent with the test's

purpose of simulating a severe fire by raising the container's

temperature and pressure by completely surrounding it with flames

produced by a specific fuel type.

F. Labeling Requirements

In the NPRM, NHTSA proposed to require that container manufacturers

certify that each of their containers complies with the proposed

equipment requirements and permanently label the container with the

following information: the symbol ``DOT'' to constitute a certification

by the manufacturer that the container conforms to all requirements of

the standard; the date of manufacture of the container; the name and

address of the container manufacturer; and the maximum service

pressure. The agency stated that labeling the container would provide

vehicle manufacturers and consumers with assurance that they are

purchasing containers that comply with the Federal safety standards. In

addition, the agency believed that the proposed requirement would

facilitate the agency's enforcement efforts by providing a ready means

of identifying the container and its manufacturer.

EDO, NGVC, Thomas, NYCFD, and Volvo GM addressed the proposed

labeling requirements. EDO and NYCFD stated that the label should

include the maximum fill pressure at a location close to the fill

receptacle. NGVC recommended that a blank area for the container

installation date be included in the label to be filled in by the

installer. Volvo GM stated that only containers that are manufactured

after the standard's effective date, and therefore actually subject to

the standard, should be entitled to display the DOT symbol as

certification of compliance with the standard. Thomas stated, without

elaboration, that the labeling requirements of NGV2 should be adopted.

NHTSA's proposal did not include certain additional information

included in NGV2, including the type of container, inspector symbols,

trademarks, manufacturer's part number, and serial numbers.

After reviewing the comments, NHTSA has decided to adopt the

proposed labeling requirements with a slight modification from the

proposed format. In item (a), the agency has modified the proposal

which states ``The tank manufacturer's name and address'' to state the

following: include the statement that ``If there is a question about

the proper use, installation, or maintenance of this container, contact

[manufacturer's name, address, and telephone number].''

The agency has decided not to require the other additional items of

information in NGV2 since the agency did not propose the inclusion of

such information in the NPRM. Notwithstanding the agency's decision not

to require this additional information, a manufacturer may list such

information on the label, provided the additional information does not

obscure or confuse the required information. In particular, NHTSA

encourages manufacturers to include the container type, e.g., Type 1,

2, 3 or 4, since the agency has decided to adopt NGV2's design and

material specifications in this final rule. Specifying the type of

container should facilitate oversight of compliance tests since each

type of container is required to undergo hydrostatic burst tests, but

with different safety factors.

In the upcoming SNPRM, NHTSA anticipates proposing additional

requirements about the CNG fuel container's label, including the

container type. In addition, the agency anticipates proposing that the

label include an additional statement addressing the container's

inspection and maintenance. Specifically, the label would state that

``This container should be visually inspected after an accident or fire

or at least every 12 months for damage and deterioration in accordance

with the applicable Compressed Gas Association guidelines.'' The agency

believes that such a statement would alert owners to the desirability

for reinspection over time or in the event of an accident. NHTSA will

also propose requirements related to the label's location, in response

to EDO's and NYCFD's comment that the maximum service pressure should

be labeled in an area close to the fill receptacle.

G. Leadtime

In the NPRM, NHTSA proposed to make the equipment requirements

effective on September 1, 1994. The agency believed that this would

provide a reasonable time period for manufacturers to make minor

modifications in container design. This proposal was based on the

agency's belief that the proposed requirements were similar to RSPA

standards currently in effect. The agency requested comment on the

feasibility of this effective date.

NHTSA received eleven comments about the proposed effective date

applicable to the container requirements. The commenters were TMC, the

U.S. Department of Energy, TBB, Oklahoma Gas, NGVC, EDO, Volvo/GM,

AAMA, ARC, Navistar, and NGV Systems. EDO and Navistar requested that

the final rule be issued as early as possible. DOE and Oklahoma Gas

recommended an effective date of September 1, 1995. NGVC recommended an

effective date of September 1, 1996, unless NGV2 were adopted which

would permit an immediate supply of containers. NGV Systems stated that

an earlier effective date would be difficult to meet since the rule, as

proposed, would require new tooling, process development, and perhaps

equipment modification. ARC stated that the rule, as proposed, would

require major modifications, since its containers have been designed to

comply with NGV2. AAMA and Volvo/GM stated that the effective dates for

the vehicle requirements and the equipment requirements should not be

concurrent.

NHTSA notes that these comments were based on the requirements, as

proposed in the NPRM. Since the final rule has been made essentially

consistent with NGV2 (with the exception of carbon fiber containers),

the agency anticipates that container manufacturers can for the most

part already certify that containers, other than carbon fiber ones,

comply with the new standard. This belief is based on comments on the

NPRM and meetings with NGVC, the CGA, and CNG container manufacturers.

With regard to manufacturers of carbon fiber containers, EDO indicated

that it already complies with the standard and Brunswick indicated that

it would need less than one month lead time for a safety factor greater

than 2.25. Brunswick further stated that it would need an unspecified

time period to modify the mounting brackets and other hardware. The CNG

industry groups have informed the agency that they want a CNG fuel

container standard to be effective as quickly as possible. In addition,

they favor having an opportunity to ``voluntarily certify compliance''

to the standard once the final rule is published. The CNG industry

groups believe that it is necessary for Federal standards to be in

place as soon as possible, given the expected increased demand for CNG

containers in light of Federal and State fleet programs for clean fuel

vehicles. They also favor quick adoption of a Federal standard to

preempt state regulations that otherwise may be promulgated and to

ensure that substandard CNG containers are not marketed.

After reviewing the comments, NHTSA has decided to establish an

effective date six months after the final rule is issued. As explained

above, most CNG containers can be certified to comply with the new

Federal motor vehicle safety standard since they already comply with

NGV2 or can be modified so that they comply within six months.

Nevertheless, the agency believes that it is necessary to provide a

leadtime of six months to allow manufacturers time to make whatever

design changes are necessary and to conduct testing so that they can

certify that their containers comply with the new standard. In the

meantime, prior to the standard's effective date, the industry is free

to advertise containers as meeting the CNG equipment standard that will

take effect in six months.\8\ Manufacturers have taken the approach of

seeking early compliance with respect to other agency requirements such

as those relating to dynamic side impact protection and air bags.

Therefore, the agency encourages manufacturers to seek, to the extent

feasible, to manufacture their CNG containers to meet these new

requirements before the date the standard takes effect.

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

\8\However, the agency emphasizes that a manufacturer may not

certify a container as meeting the equipment standard until the

standard goes into effect. Under the Vehicle Safety Act, a

certification is a statement that a vehicle or item of equipment

meets all applicable Federal Motor Vehicle Safety Standards that are

then in effect. Therefore, until a standard is effective,

manufacturers may not certify compliance with it.

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

With regard to the concern expressed by AAMA and Volvo GM that the

effective date of the container regulation should precede that of the

vehicle regulation, AAMA based its comments on the belief that it will

need to know the performance of the containers it will use in the fuel

systems of its vehicles. NHTSA notes that CNG containers now typically

meet NGV2 and thus should comply with NHTSA's standards. Therefore,

AAMA members already have access to and detailed knowledge about

containers that should meet the new requirements.

H. Benefits

In the NPRM, NHTSA addressed the proposal's benefits with respect

to CNG vehicles. The notice did not directly address the benefits of

regulating the CNG fuel containers.

NHTSA received no comments directly addressing the benefits of

regulating CNG containers. Brunswick criticized the proposal, believing

that it would place carbon fiber containers at a competitive

disadvantage. Brunswick stated that the proposed single burst factor

would provide less benefits than if the agency adopted NGV2.\9\

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

\9\Because NHTSA is adopting Brunswick's request for multiple

safety factors, that commenter's concern about a single safety

factor is moot.

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

NHTSA anticipates that the number of CNG fuel vehicles will

increase greatly in the near future, in light of directives by the

Clinton Administration\10\ and legislation by Congress to develop

vehicles powered by cleaner burning fuels. This final rule will

increase the safety of this growing population of vehicles.

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

\10\Executive Order 12844 increased by 50 percent the number of

alternatively fueled vehicles to be acquired by the Federal

Government from 1993 through 1995. (April 21, 1993) In addition, in

1993, the President established the Federal Fleet Conversion Task

Force to accelerate the commercialization and market acceptance of

alternative fueled vehicles throughout the country.

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

I. Costs

In the NPRM, NHTSA stated that each container would cost $600. The

agency further stated that the container testing costs would range from

approximately $4,050 to $8,600 for each model of container.

NGVC, NGV Systems, PST, Brunswick, ARC, Thomas Built, and Flxible

addressed the costs of the proposal with respect to CNG containers.

NGVC and the CNG container manufacturers stated that the proposal,

especially given the single safety factor in the burst test

requirements, significantly understated the costs of the rulemaking.

Brunswick stated that container manufacturers would incur significant

costs since they would have to redesign and requalify their currently

designed tanks. As a result, it believed that the CNG containers would

be more expensive and heavier. It estimated that the proposal would

increase costs between 10 percent and 55 percent, depending on the

material and method of construction. Brunswick further stated that this

proposal would add many millions of dollars on an industry-wide basis.

NGVC commented that the qualification tests could cost $20,000 for

each model of container since many tests will be required on prototype

containers. It stated that some manufacturers estimate that the design,

manufacture, and qualification costs could approach $150,000 per

container model, a figure that greatly exceeded NHTSA's estimate of

$74,000.

NHTSA believes that the basis for the comments about the costs of

this rulemaking have been largely eliminated except in connection with

carbon fiber tanks. The comments were based on the proposal for a

single safety factor of 3.5 for all types of tanks. As noted above, the

agency has decided to specify multiple safety factors that are

consistent with NGV2 except in the case of the factors for carbon fiber

containers. Since all the container manufacturers commenting on the

proposal either already certify to or can comply with NGV2 without any

design changes, the cost to manufacturers will be minimal for noncarbon

fiber tanks.

V. Rulemaking Analyses

A. Executive Order 12866 and DOT Regulatory Policies and Procedures

NHTSA has considered the impact of this rulemaking action under

Executive Order 12866 and the Department of Transportation's regulatory

policies and procedures. This rulemaking document was not reviewed

under E.O. 12866, ``Regulatory Planning and Review.'' This action has

been determined to be ``nonsignificant'' under the Department of

Transportation's regulatory policies and procedures. NHTSA has

estimated the costs of the amendments in a Final Regulatory Evaluation

(FRE) which is included in the docket for this rulemaking. As discussed

in that document, NHTSA estimates that the cost for the pressure

cycling, burst, and bonfire testing will range from $9,000 to $21,725

per container size and type. In addition, the cost of the containers

used in the test is estimated to range from $1,800 to $6,600. Since the

safety factors in the burst test applicable to carbon fiber containers

are more stringent than those in NGV2, the cost of those containers

will increase. Based on comments by Brunswick and other information,

the switch from carbon fiber containers meeting a 2.25 safety factor to

carbon fiber containers meeting the factors adopted in this final rule

will increase the container cost and the lifetime fuel costs about 8.75

percent for vehicles equipped with Type 2 containers. Those costs would

be range from $115 for passenger cars to $602 for heavy trucks. The

switch would increase costs about 37.1 percent for vehicles equipped

with Type 3 and Type 4 containers, resulting in a cost increase ranging

from $496 for cars to $2,560 for heavy trucks.

B. Regulatory Flexibility Act

NHTSA has also considered the effects of this rulemaking action

under the Regulatory Flexibility Act. Based upon the agency's

evaluation, I certify that this rule will not have a significant

economic impact on a substantial number of small entities. Information

available to the agency indicates that businesses manufacturing CNG

fuel containers are not small businesses.

C. Executive Order 12612 (Federalism)

NHTSA has analyzed this rulemaking action in accordance with the

principles and criteria contained in Executive Order 12612. NHTSA has

determined that the rule will not have sufficient Federalism

implications to warrant the preparation of a Federalism Assessment. No

state has adopted requirements regulating CNG containers.

D. National Environmental Policy Act

In accordance with the National Environmental Policy Act of 1969,

NHTSA has considered the environmental impacts of this rule. The agency

has determined that this rule will have no adverse impact on the

quality of the human environment. On the contrary, because NHTSA

anticipates that ensuring the safety of CNG vehicles will encourage

their use, NHTSA believes that the rule will have positive

environmental impacts. CNG vehicles are expected to have near-zero

evaporative emissions and the potential to produce very low exhaust

emissions as well.

E. Civil Justice Reform

This final rule does not have any retroactive effect. Under 49

U.S.C. 30103, whenever a Federal motor vehicle safety standard is in

effect, a State may not adopt or maintain a safety standard applicable

to the same aspect of performance which is not identical to the Federal

standard, except to the extent that the State requirement imposes a

higher level of performance and applies only to vehicles procured for

the State's use. 49 U.S.C. 30161 sets forth a procedure for judicial

review of final rules establishing, amending or revoking Federal motor

vehicle safety standards. That section does not require submission of a

petition for reconsideration or other administrative proceedings before

parties may file suit in court.

List of Subjects in 49 CFR Part 571

Imports, Incorporation by reference, Motor vehicle safety, Motor

vehicles.

PART 571--[AMENDED]

In consideration of the foregoing, 49 CFR Part 571 is amended as

follows:

PART 571--[AMENDED]

1. The authority citation for Part 571 continues to read as

follows:

Authority: 49 U.S.C. 322, 30111, 30115, 30117 and 30166;

delegation of authority at 49 CFR 1.50.

2. Section 571.5 is amended by redesignating (b)(7) as (b)(10) and

adding new paragraphs (b)(7) through (b)(9), to read as follows:

Sec. 571.5 Matter incorporated by reference.

* * * * *

(b) * * *

(7) Standards of Suppliers of Advanced Composite Materials

Association (SACMA). They are published by Suppliers of Advanced

Composite Materials Association. Information and copies may be obtained

by writing to: Suppliers of Advanced Composite Materials Association,

1600 Wilson Blvd., Suite 1008, Arlington, VA 22209.

(8) Standards of the American Society of Mechanical Engineers

(ASME). They are published by The American Society of Mechanical

Engineers. Information and copies may be obtained by writing to: The

American Society of Mechanical Engineers, 345 East 47th Street, New

York, NY 10017.

(9) Computer Analysis by the National Aeronautics and Space

Administration (NASA). This was conducted by the National Aeronautics

and Space Administration. Information and copies may be obtained by

writing to: National Aeronautics and Space Administration, 600

Independence Avenue SW, Washington, DC 20546.

* * * * *

3. A new Sec. 571.304, Standard No. 304; Compressed Natural Gas

Fuel Container Integrity, is added to Part 571, to read as follows:

Sec. 571.304 Standard No. 304; Compressed Natural Gas Fuel Container

Integrity.

S1. Scope. This standard specifies requirements for the integrity

of compressed natural gas (CNG), motor vehicle fuel containers.

S2. Purpose. The purpose of this standard is to reduce deaths and

injuries occurring from fires that result from fuel leakage during and

after motor vehicle crashes.

S3. Application. This standard applies to containers designed to

store CNG as motor fuel on-board any motor vehicle.

S4. Definitions.

Brazing means a group of welding processes wherein coalescence is

produced by heating to a suitable temperature above 800 deg.F and by

using a nonferrous filler metal, having a melting point below that to

the base metals. The filler metal is distributed between the closely

fitted surfaces of the joint by capillary attraction.

Burst pressure means the highest internal pressure reached in a CNG

fuel container during a burst test at a temperature of 21 deg.C (70

deg.F).

CNG fuel container means a container designed to store CNG as motor

fuel on-board a motor vehicle.

Fill pressure means the internal pressure of a CNG fuel container

attained at the time of filling. Fill pressure varies according to the

gas temperature in the container which is dependent on the charging

parameters and the ambient conditions.

Full wrapped means applying the reinforcement of a filament or

resin system over the entire liner, including the domes.

Hoop wrapped means winding of filament in a substantially

circumferential pattern over the cylindrical portion of the liner so

that the filament does not transmit any significant stresses in a

direction parallel to the cylinder longitudinal axis.

Hydrostatic pressure means the internal pressure to which a CNG

fuel container is taken during testing set forth in S5.4.1.

Liner means the inner gas tight container or gas cylinder to which

the overwrap is applied.

Service pressure means the internal settled pressure of a CNG fuel

container at a uniform gas temperature of 21 deg.C (70 deg.F) and

full gas content. It is the pressure for which the container has been

constructed under normal conditions.

Stress ratio means the stress in the fiber at minimum burst

pressure divided by the stress in the fiber at service pressure.

S5 Container and material requirements.

S5.1 Container designations. Container designations are as

follows:

S5.1.1 Type 1--Non-composite metallic container means a metal

container.

S5.1.2 Type 2--Composite metallic hoop wrapped container means a

metal liner reinforced with resin impregnated continuous filament that

is ``hoop wrapped.''

S5.1.3 Type 3--Composite metallic full wrapped container means a

metal liner reinforced with resin impregnated continuous filament that

is ``full wrapped.''

S5.1.4 Type 4--Composite non-metallic full wrapped container means

resin impregnated continuous filament with a non-metallic liner ``full

wrapped.''

S5.2 Material designations.

S5.2.1 Steel containers and liners.

(a) Steel containers and liners shall be of uniform quality. Only

the basic oxygen or electric furnace processes are authorized. The

steel shall be aluminum killed and produced to predominantly fine grain

practice. The steel heat analysis shall be in conformance with one of

the following grades:

Table One--Steel Heat Analysis

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

Chrome-Molybdenum Carbon-Manganese

Grade element percent Carbon-Boron percent percent

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

Carbon..................................... 0.25 to 0.38......... 0.27 to 0.37......... 0.40 max.

Manganese.................................. 0.40 to 1.05......... 0.80 to 1.40......... 1.65 max.

Phosphorus................................. 0.015 max............ 0.015 max............ 0.025 max.

Sulfur..................................... 0.010 max............ 0.010 max............ 0.010 max.

Silicon.................................... 0.15 to 0.35......... 0.30 max............. 0.10/0.30

Chromium................................... 0.80 to 1.15......... N/A.................. N/A

Molybdenum................................. 0.15 to 0.25......... N/A.................. N/A

Boron...................................... N/A.................. 0.0005 to 0.003...... N/A

Aluminum................................... 0.02 to 0.07......... 0.02 to 0.07......... 0.02/0.07

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

\1\``N/A'' means not applicable.

(b) Incidental elements shall be within the limits specified in the

Standard Specification for Steel, Sheet and Strip, Alloy, Hot-Rolled

and Cold-Rolled, General Requirements for ASTM A 505 (1987).

S5.2.1.1 When carbon-boron steel is used, the test specimen is

subject to a hardenability test in accordance with the Standard Method

for End-Quench Test For Hardenability of Steel, ASTM A 255 (1989). The

hardness evaluation is made 7.9 mm (\5/16\ inch) from the quenched end

of the Jominy quench bar.

S5.2.1.2 The test specimen's hardness shall be at least Rc

(Rockwell Hardness) 33 and no more than Rc 53.

S5.2.2 Aluminum containers and aluminum liners. (Type 1, Type 2

and Type 3) shall be 6010 alloy, 6061 alloy, and T6 temper. The

aluminum heat analysis shall be in conformance with one of the

following grades:

Table Two--Aluminum Heat Analysis

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

6061 alloy

Grade element 6010 alloy percent percent

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

Magnesium......................... 0.60 to 1.00...... 0.60 to 1.20

Silicon........................... 0.80 to 1.20...... 0.40 to 0.80

Copper............................ 0.15 to 0.60...... 0.15 to 0.40

Chromium.......................... 0.05 to 0.10...... 0.04 to 0.35

Iron.............................. 0.50 max.......... 0.70 max.

Titanium.......................... 0.10 max.......... 0.15 max.

Manganese......................... 0.20 to 0.80...... 0.15 max.

Zinc.............................. 0.25 max.......... 0.25 max.

Bismuth........................... 0.003 max......... 0.003 max.

Lead.............................. 0.003 max......... 0.003 max.

Others, Each\1\................... 0.05 max.......... 0.05 max.

Others, Total\1\.................. 0.15 max.......... 0.15 max.

Aluminum.......................... Remainder......... Remainder.

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

\1\Analysis is made only for the elements for which specific limits are

shown, except for unalloyed aluminum. If, however, the presence of

other elements is indicated to be in excess of specified limits,

further analysis is made to determine that these other elements are

not in excess of the amount specified. (Aluminum Association Standards

and Data--Sixth Edition 1979.)

S5.2.3 Structural reinforcing filament material shall be

commercial grade E-glass, commercial grade S-glass, aramid fiber or

carbon fiber. Filament strength shall be tested in accordance with the

Standard Test Method for Tensile Properties of Glass Fiber Strands,

Yarns, and Rovings Used in Reinforced Plastics, ASTM D 2343 (1967,

Reapproved 1985), or SACMA Recommended Test Method for Tow Tensile

Testing of Carbon Fibers, SRM 16-90, 1990. Fiber coupling agents

(sizing) shall be compatible with the resin system. If carbon fiber

reinforcement is used the design shall incorporate means to prevent

galvanic corrosion of metallic components of the fuel container.

S5.2.4 The resin system shall be epoxy, modified epoxy, polyester,

vinyl ester or thermoplastic.

S5.2.4.1 The resin system is tested on a sample coupon

representative of the composite overwrap in accordance with the

Standard Test Method for Apparent Interlaminar Shear Strength of

Parallel Fiber Composites by Short-Beam Method, ASTM D 2344, (1984,

Reapproved 1989) following a 24-hour water boil.

S5.2.4.2 The test specimen shall have a shear strength of at least

13.8 MPa (2,000 psi).

S5.2.5 For nonmetallic liners, the permeation of CNG through the

finished container's wall at service pressure is less than 0.25 normal

cubic centimeters per hour per liter water capacity of the container.

S5.3 Manufacturing processes for composite containers.

S5.3.1 Composite containers with metallic liners. The CNG fuel

container shall be manufactured from a metal liner overwrapped with

resin impregnated continuous filament windings, applied under

controlled tension to develop the design composite thickness. After

winding is complete, composites using thermoset resins shall be cured

by a controlled temperature process.

S5.3.1.1 Type 2 containers. Type 2 containers shall have a hoop

wrapped winding pattern.

S5.3.1.2 Type 3 containers. Type 3 containers shall have a full

wrapped ``helical or in plane'' and a ``hoop'' wrap winding pattern.

S5.3.2 Type 4 containers. Composite containers with nonmetallic

liners shall be fabricated from a nonmetallic liner overwrapped with

resin impregnated continuous filament windings. The winding pattern

shall be ``helical or in plane'' and ``hoop'' wrap applied pattern

under controlled tension to develop the design composite thickness.

After winding is complete, the composite shall be cured by a controlled

temperature process.

S5.3.3 Brazing. Brazing is prohibited.

S5.3.4 Welding. Welding shall be done in accordance with the

American Society of Mechanical Engineers (ASME) Boiler and Pressure

Vessel Code, Section IX, Article II, QW-304 and QW-305 (1992). Weld

efficiencies shall be in accordance with ASME Boiler and Pressure

Vessel Code, Section VIII, UW-12 (1989). Any weld shall be subject to

full radiographic requirements in accordance with ASME Boiler and

Pressure Vessel Code, Section VIII, UW-51 thru UW-53 (1989). For Type 2

and Type 3 liners, longitudinal welds and nonconsumable backing strips

or rings shall be prohibited.

S5.4 Wall thickness.

S5.4.1 Type 1 containers.

(a) The wall thickness of a Type 1 container shall be at least an

amount such that the wall stress at the minimum prescribed hydrostatic

test pressure does not exceed 67 percent of the minimum tensile

strength of the metal as determined by the mechanical properties

specified in S5.7 and S5.7.1.

(b) For minimum wall thickness calculations, the following formula

is used:

TR26SE94.000

Where:

S = Wall stress in MPa (psi).

P = Minimum hydrostatic test pressure in Bar (psig).

D = Outside diameter in mm (inches).

d = Inside diameter in mm (inches).

S5.4.2 Type 2 containers.

S5.4.2.1 The wall thickness of a liner to a Type 2 container shall

be at least an amount such that the longitudinal tensile stress at the

minimum design burst pressure does not exceed the ultimate tensile

strength of the liner material as determined in S5.7 and S5.7.1.

S5.4.2.2 The wall thickness of a liner to a Type 2 container

shall be at least an amount such that the compressive stress in the

sidewall of the finished container at zero pressure shall not exceed 95

percent of the yield strength of the liner as determined in S5.7 and

S5.7.1 or 95 percent of the minimum design yield strength shown in

S5.7.3. The maximum tensile stress in the liner at service pressure

shall not exceed 66 percent of the yield strength.

S5.4.2.3 Stresses at the end designs at internal pressures between

no more than 10 percent of service pressure and service pressure shall

be less than the maximum stress limits in the sidewall as prescribed

above.

S5.4.3 Type 3 containers. The wall thickness of a liner to a Type

3 container shall be such that the compressive stress in the sidewall

of the finished container at zero pressure shall not exceed 95 percent

of the minimum yield strength of the liner as determined in S5.7 and

S5.7.1 or 95 percent of the minimum design yield strength shown in

S5.7.3

S5.4.4 Type 4 containers. The wall thickness of a liner to a Type

4 container shall be such that the permeation rate requirements of this

specification are met.

S5.5 Composite reinforcement for Type 2, Type 3, and Type 4

Containers.

S5.5.1 Compute stresses in the liner and composite reinforcement

using National Aeronautics and Space Administration (NASA) NAS 3-6292,

Computer Program for the Analysis of Filament Reinforced Metal-Shell

Pressure Vessels, (May 1966).

S5.5.2 The composite overwrap shall meet or exceed the following

composite reinforcement stress ratio values shown in Table 3.

S5.6 Thermal treatment.

S5.6.1 Steel containers or liners.

S5.6.1.1 After all metal forming and welding operations, completed

containers or liners shall be uniformly and properly heat treated under

the same conditions of time, temperature and atmosphere prior to all

tests.

S5.6.1.2 All containers or liners of steel grades ``Chrome-

Molybdenum'' or ``Carbon Boron'' shall be quenched in a medium having a

cooling rate not in excess of 80 percent that of water. ``Carbon-

Manganese'' steel grades shall be normalized and do not require

tempering after normalizing.

S5.6.1.3 All steel temperature on quenching shall not exceed

926 deg.C (1700 deg.F).

S5.6.1.4 All containers or liners or steel grades ``Chrome-

Molybdenum'' or ``Carbon Boron'' shall be tempered after quenching at a

temperature below the transformation ranges, but not less than

482 deg.C (900 deg.F) for ``Carbon-Boron'' steel or 565 deg.C

(1050 deg.F) for ``Chrome-Molybdenum'' steel. ``Carbon Manganese''

steel grades do not require tempering after normalizing.

S5.6.2 Aluminum containers or liners (seamless and welded). After

all forming and welding operations, aluminum containers or liners shall

be solution heat treated and aged to the T6 temper. The liner and

composite overwrap shall meet the cycle life and strength requirements

set forth in S7.1 and S7.2 of this standard.

S5.7 Yield strength, tensile strength, material elongation (metal

containers and metal liners only). To determine yield strength, tensile

strength, and elongation of the material, cut two specimens from one

container or liner. The specimen either has (a) a gauge length of 50 mm

(2 inches) and a width not over 38 mm (1.5 inches), or (b) a gauge

length of four times the specimen diameter, provided that a gauge

length which is at least 24 times the thickness with a width not over 6

times the thickness is permitted when the liner wall is not over 5 mm

(3/16 inch) thick. The specimen shall not be flattened, except that

grip ends may be flattened to within 25 mm (1 inch) of each end of the

reduced section. Heating of specimens is prohibited.

S5.7.1 Yield strength. The yield strength in tension shall be the

stress corresponding to a permanent strain of 0.2 percent based on the

gauge length.

S5.7.1.1 The yield strength shall be determined by either the

``offset'' method or the ``extension under load'' method as prescribed

by Standard Test Methods for Tension Testing of Metallic Materials,

ASTM E8 1993.

S5.7.1.2 In using the ``extension under load'' method, the total

strain or ``extension under load'' corresponding to the stress at which

the 0.2 percent permanent strain occurs may be determined by

calculating the elastic extension of the gauge length under appropriate

load and adding thereto 0.2 percent of the gauge length. Elastic

extension calculations shall be based on an elastic modulus of 69 GPa

(10,000,000 psi) for aluminum, or 207 GPa (30,000,000 psi) for steel.

If the elastic extension calculation does not provide a conclusive

result, the entire stress strain diagram shall be plotted and the yield

strength determined from the 0.2 percent offset.

S5.7.1.3 For the purpose of strain measurement, the initial strain

is set while the test specimen is under a stress of 41 MPa (6,000 psi)

for aluminum, and 83 MPa (12,000 psi) for steel. The strain indicator

reading is set at the calculated corresponding strain.

S5.7.1.4 Cross-head speed of the testing machine is 3.2 mm (1/8

inch) per minute or less during yield strength determination.

S5.7.2 Elongation. Elongation of material, when tested in

accordance with S5.7, shall be at least 14 percent for aluminum or at

least 20 percent for steel; except that an elongation of 10 percent is

acceptable for both aluminum and steel when the authorized specimen

size is 24t gauge length x 6t wide, where ``t'' equals specimen

thickness.

S5.7.3 Tensile strength. Tensile strength shall not exceed 725 MPa

(105,000 psi) for ``Carbon Manganese'' and 966 MPa (140,000 psi) for

``Chrome-Molybdenum'' and ``Carbon-Boron.''

S6 General requirements.

S6.1 Each passenger car, multipurpose passenger vehicle, truck,

and bus that uses CNG as a motor fuel shall be equipped with a CNG fuel

container that meets the requirements of S7 through S7.4.

S6.2 Each CNG fuel container manufactured on or after March 27,

1994, shall meet the requirements of S7 through S7.4.

S7 Test requirements. Each CNG fuel container shall meet the

applicable requirements of S7 through S7.4.

S7.1 Pressure cycling test at ambient temperature. Each CNG fuel

container shall not leak when tested in accordance with S8.1.

S7.2 Hydrostatic burst test.

S7.2.1 Each Type 1 CNG fuel container shall not leak when

subjected to burst pressure and tested in accordance with S8.2. Burst

pressure shall be not less than 2.25 times the service pressure for

non-welded containers when analyzed in accordance with the stress ratio

requirements of S5.4.1, and shall not be less than 3.5 times the

service pressure for welded containers.

S7.2.2 Each Type 2, Type 3, or Type 4 CNG fuel container shall not

leak when subjected to burst pressure and tested in accordance with

S8.2. Burst pressure shall be no less than the value necessary to meet

the stress ratio requirements of Table 3, when analyzed in accordance

with the requirements of S5.5.1. Burst pressure is calculated by

multiplying the service pressure by the applicable stress ratio set

forth in Table Three.

Table Three--Stress Ratios

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

Material Type 2 Type 3 Type 4

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

E-Glass...................................... 2.65 3.5 3.5

S-Glass...................................... 2.65 3.5 3.5

Aramid....................................... 2.25 3.0 3.0

Carbon....................................... 2.50 3.33 3.33

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

S7.3 Bonfire test. Each CNG fuel container shall be equipped with

a pressure relief device. Each CNG fuel container shall completely vent

its contents through a pressure relief device or shall not burst while

retaining its entire contents when tested in accordance with S8.3.

S7.4. Labeling. Each CNG fuel container shall be permanently

labeled with the information specified in paragraphs (a) through (d).

The information specified in paragraphs (a) through (d) of this section

shall be in English and in letters and numbers that are at least 12.7

mm (\1/2\ inch) high.

(a) The statement: ``If there is a question about the proper use,

installation, or maintenance of this container, contact

________________.'' inserting the CNG fuel container manufacturer's

name, address, and telephone number.

(b) The statement: ``Manufactured in ____________.'' inserting the

month and year of manufacture of the CNG fuel container.

(c) Maximum service pressure ________ kPa (________ psig).

(d) The symbol DOT, constituting a certification by the CNG

container manufacturer that the container complies with all

requirements of this standard.

S8 Test conditions: fuel container integrity.

S8.1 Pressure cycling test. The requirements of S7.1 shall be met

under the conditions of S8.1.1 through S8.1.4.

S8.1.1 Hydrostatically pressurize the CNG container to the service

pressure, then to not more than 10 percent of the service pressure, for

13,000 cycles.

S8.1.2 After being pressurized as specified in S8.1.1,

hydrostatically pressurize the CNG container to 125 percent of the

service pressure, then to not more than 10 percent of the service

pressure, for 5,000 cycles.

S8.1.3 The cycling rate for S8.1.1 and S8.1.2 shall not exceed 10

cycles per minute.

S8.1.4 The cycling is conducted at ambient temperature.

S8.2 Hydrostatic burst test. The requirements of S7.2 shall be met

under the conditions of S8.2.1 through S8.2.2.

S8.2.1 Hydrostatically pressurize the CNG fuel container, as

follows: The pressure is increased up to the minimum prescribed burst

pressure determined in S7.2.1 or S7.2.2, and held constant at the

minimum burst pressure for 10 seconds.

S8.2.2 The pressurization rate throughout the test shall not

exceed 1,379 kPa (200 psi) per second.

S8.3 Bonfire test. The requirements of S7.3 shall be met under the

conditions of S8.3.1 through S8.3.10.

S8.3.1 The CNG fuel container is filled with compressed natural

gas and tested at (1) 100 percent of service pressure and (2) 25

percent of service pressure. Manufacturers may conduct these tests

using the same container or with separate containers.

S8.3.2 The CNG fuel container is positioned so that its

longitudinal axis is horizontal. Subject the entire length to flame

impingement, except that the flame shall not be allowed to impinge

directly on any pressure relief device. Shield the pressure relief

device with a metal plate.

S8.3.3 If the test container is 165 cm (65 inches) in length or

less, place it in the upright position and subject it to total fire

engulfment in the vertical. The flame shall not be allowed to impinge

directly on any pressure relief device. For containers equipped with a

pressure relief device on one end, the container is positioned with the

relief device on top. For containers equipped with pressure relief

devices on both ends, the bottom pressure relief device shall be

shielded with a metal plate.

S8.3.4 The lowest part of the container is 102 mm (4 inches) above

the liquid surface of the diesel fuel at the beginning of the test.

S8.3.5 The CNG fuel container is tested with the valve and

pressure relief device or devices in place.

S8.3.6 The fire is generated by No. 2 diesel fuel.

S8.3.7 The fuel specified in S8.3.6 is contained in a pan such

that there is sufficient fuel to burn for at least 20 minutes. The

pan's dimensions ensure that the sides of the fuel containers are

exposed to the flame. The pan's length and width shall exceed the fuel

container projection on a horizontal plane by at least 20 cm (8 inches)

but not more than 50 cm (20 inches). The pan's sidewalls shall not

project more than 2 cm (0.8 inches) above the level of fuel.

S8.3.8 Time-pressure readings are recorded at 30 second intervals,

beginning when the fire is lighted and continuing until the container

is completely tested.

S8.3.9 The CNG fuel container is exposed to the bonfire for 20

minutes or until its contents are completely vented.

S8.3.10 The average wind velocity at the container is not to

exceed 2.24 meters/second (5 mph).

Issued on September 16, 1994.

Ricardo Martinez,

Administrator.

[FR Doc. 94-23571 Filed 9-21-94; 1:13 pm]

BILLING CODE 4910-50-P