ISO/TR 12391-2:2002

Gas cylinders — Refillable seamless steel — Performance tests — Part 2: Fracture performance tests — Monotonic burst tests

This part of ISO/TR 12391 is a summary and compilation of the test results obtained during the development of the "Flawed-Cylinder Burst Test". The concept and development of the flawed cylinder burst test is described in ISO/TR 12391-1. The test is a method for evaluating the fracture performance of steel cylinders that are used to transport high pressure, compressed gases. In this part of ISO/TR 12391, test results are reported for several hundred flawed cylinder burst tests that were conducted on seamless steel cylinders ranging in tensile strength from less than 750 MPa up to about 1 400 MPa. This test method has been shown to reliably predict the fracture performance of seamless steel cylinders. The test method is intended to be used both for the selection of materials and design parameters in the development of new cylinder designs as well as for an efficient quality control test to be used during the production of cylinders.

Bouteilles à gaz — Rechargeables en acier sans soudure — Essais de performance — Partie 2: Essais de mode de rupture — Essais de rupture monotonique

General Information

• Status: Published
• Publication Date: 10-Dec-2002
• ICS: 23.020.30 - Pressure vessels, gas cylinders
• ICS: 23.020.35 - Gas cylinders
• Technical Committee: ISO/TC 58/SC 3 - Cylinder design
• Drafting Committee: ISO/TC 58/SC 3 - Cylinder design
• Current Stage: 9093 - International Standard confirmed
• Completion Date: 27-Oct-2017

Standards Content (Sample)

TECHNICAL ISO/TR
REPORT 12391-2
First edition
2002-12-15

Gas cylinders — Refillable seamless
steel — Performance tests —
Part 2:
Fracture performance tests — Monotonic
burst tests
Bouteilles à gaz — Rechargeables en acier sans soudure — Essais de
performance —
Partie 2: Essais de mode de rupture — Essais de rupture monotonique




Reference number
ISO/TR 12391-2:2002(E)
©
ISO 2002

---------------------- Page: 1 ----------------------
ISO/TR 12391-2:2002(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.


©  ISO 2002
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail [e-mail]
Web www.iso.org
Published in Switzerland

ii © ISO 2002 — All rights reserved

---------------------- Page: 2 ----------------------
ISO/TR 12391-2:2002(E)
Contents Page
Foreword. iv
Introduction . v
1 Scope. 1
2 References. 1
3 Symbols . 2
4 Background information . 2
5 Experimental test programme . 4
5.1 Types of cylinder tested. 4
5.2 Material properties tests. 6
5.3 Description of the flawed-cylinder burst test. 7
6 Flawed-cylinder burst test results. 9
6.1 Flawed-cylinder burst test procedure. 9
6.2 Flawed-cylinder burst test results for group A materials. 10
6.3 Flawed-cylinder burst test results for group B materials. 11
6.4 Flawed-cylinder burst test results for group C materials. 11
6.5 Flawed-cylinder burst test results for group D materials. 11
6.6 Flawed-cylinder burst test results for group E materials . 12
6.7 Flawed-cylinder burst test results for tests conducted under special conditions . 12
7 Discussion . 12
7.1 Background . 12
7.2 ISO 9809-2 flawed-cylinder burst test procedures and acceptance criteria. 13
7.3 Analysis by WG 14 to relate the flawed-cylinder burst test to Charpy-V-notch energy
values . 14
7.4 Adjustment to the measured P /P ratio to account for the local cylinder wall thickness . 15
f s
8 Summary and conclusions . 16
Annex A (informative) Evaluation of the measurement uncertainty in the flawed-cylinder burst
test. 113
Bibliography . 116

© ISO 2002 — All rights reserved iii

---------------------- Page: 3 ----------------------
ISO/TR 12391-2:2002(E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
In exceptional circumstances, when a technical committee has collected data of a different kind from that
which is normally published as an International Standard (“state of the art”, for example), it may decide by a
simple majority vote of its participating members to publish a Technical Report. A Technical Report is entirely
informative in nature and does not have to be reviewed until the data it provides are considered to be no
longer valid or useful.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO/TR 12391-2 was prepared by Technical Committee ISO/TC 58, Gas cylinders, Subcommittee SC 3,
Cylinder design.
ISO/TR 12391 consists of the following parts, under the general title Gas cylinders — Refillable seamless
steel — Performance tests:
 Part 1: Philosophy, background and conclusions
 Part 2: Fracture performance tests — Monotonic burst tests
 Part 3: Fracture performance tests — Cyclical burst tests
 Part 4: Flawed-cylinder cycle test
iv © ISO 2002 — All rights reserved

---------------------- Page: 4 ----------------------
ISO/TR 12391-2:2002(E)
Introduction
Gas cylinders as specified in ISO 9809-1 have been constructed of steel with a maximum tensile strength of
less than 1 100 MPa. With the technical changes in steel-making using a two-stage process, referred to as
ladle metallurgy or secondary refining, significant improvement in mechanical properties have been achieved.
These improved mechanical properties provide the opportunity of producing gas cylinders with higher tensile
strength and which achieve a lower ratio of steel weight to gas weight. The major concern in using steels of
higher tensile strength with correspondingly higher design wall stress is safety throughout the life of the gas
cylinder.
When ISO/TC 58/SC 3 began drafting ISO 9809-2, Working Group 14 was formed to study the need for
additional controls for the manufacture of steel gas cylinders having a tensile strength greater than
1 100 MPa.
This part of ISO/TR 12391 presents all of the specific test results of the monotonic, flawed-cylinder burst tests
that were conducted in order to evaluate the fracture performance of cylinders ranging in tensile strength from
less 750 MPa to greater than 1 210 MPa.

© ISO 2002 — All rights reserved v

---------------------- Page: 5 ----------------------
TECHNICAL REPORT ISO/TR 12391-2:2002(E)

Gas cylinders — Refillable seamless steel — Performance
tests —
Part 2:
Fracture performance tests — Monotonic burst tests
1 Scope
This part of ISO/TR 12391 is a summary and compilation of the test results obtained during the development
of the “Flawed-Cylinder Burst Test”. The concept and development of the flawed cylinder burst test is
described in ISO/TR 12391-1. The test is a method for evaluating the fracture performance of steel cylinders
that are used to transport high pressure, compressed gases. In this part of ISO/TR 12391, test results are
reported for several hundred flawed cylinder burst tests that were conducted on seamless steel cylinders
ranging in tensile strength from less than 750 MPa up to about 1 400 MPa.
This test method has been shown to reliably predict the fracture performance of seamless steel cylinders. The
test method is intended to be used both for the selection of materials and design parameters in the
development of new cylinder designs as well as for an efficient quality control test to be used during the
production of cylinders.
2 References
ISO 148:1983, Steel — Charpy impact test (V-notch)
ISO 6892:1998, Metallic materials — Tensile testing at ambient temperature
ISO 9809-1:1999, Gas cylinders — Refillable seamless steel gas cylinders — Design, construction and
testing — Part 1: Quenched and tempered steel cylinders with tensile strength less than 1 100 MPa
ISO 9809-2:2000, Gas cylinders — Refillable seamless steel gas cylinders — Design, construction and
testing — Part 2: Quenched and tempered steel cylinders with tensile strength greater than or equal to
1 100 MPa
ISO/TR 12391-1, Gas cylinders — Refillable seamless steel — Performance tests — Part 1: Philosophy,
background and conclusions
ISO/TR 12391-3, Gas cylinders — Refillable seamless steel — Performance tests — Part 3: Fracture
performance tests — Cyclical burst tests
© ISO 2002 — All rights reserved 1

---------------------- Page: 6 ----------------------
ISO/TR 12391-2:2002(E)
3 Symbols
A is the elongation expressed as a percentage (= d/t );
d
d is the flaw depth, expressed in millimetres (= A × t );
d
D is the outside diameter of the cylinder, expressed in millimetres;
l is the flaw length, expressed in millimetres (= n × t );
o d
n represents multiples of t (= l /t );
d o d
P is the failure pressure measured in the flawed-cylinder burst test expressed in bar.
f
P is the calculated design test pressure for the cylinder, expressed in bar;
h
P is the calculated design service pressure for the cylinder, expressed in bar;
s
R is the actual measured value of yield strength, expressed in megapascals;
ea
R is the maximum value of tensile strength guaranteed by the manufacturer, expressed in megapascals;
g, max
R is the minimum value of tensile strength guaranteed by the manufacturer, expressed in megapascals;
g, min
R is the actual measured value of tensile strength expressed in megapascals;
m
t is the actual measured wall thickness at the location of the flaw, expressed in millimetres;
a
t is the calculated minimum design wall thickness, expressed in millimetres.
d
4 Background information
High-pressure industrial gases (such as oxygen, nitrogen, argon, hydrogen, helium) are stored and
transported in portable steel cylinders. These cylinders are designed, manufactured and maintained in
accordance with ISO 9809-1, ISO 9809-2, or national specifications such as those of the U.S. Department of
[1]
Transportation (DOT) 49 CFR Part 178 . The cylinders are constructed from specified alloy steels that are
[2]
generally modified versions of steels such as AISI 4130 and AISI 4140 or equivalent steels made to other
national specifications. The cylinders are of seamless construction and are manufactured by either a forging
process, a tube drawing process or a plate drawing process. The required mechanical properties are obtained
by using an austenitizing, quenching and tempering heat treatment. Typical sizes of these cylinders are
100 mm to 250 mm in diameter, 500 mm to 2 000 mm in length, and 3 mm to 20 mm in wall thickness. Typical
working pressure ranges are 100 bar to 400 bar.
Until recently, the tensile strength of the steels used in the construction of such cylinders has been limited to a
maximum of about 1 100 MPa. This limitation for the maximum tensile strength occurs because the fracture
toughness of the steels decreases with increase in the tensile strength, and above a tensile strength of about
1 100 MPa the fracture toughness was not adequate to prevent fracture of the cylinders. Recently developed
new alloy steels that are modifications of the AISI 4130 and AISI 4140, and which have both high tensile
strength and high fracture toughness make it possible to construct lighter cylinders with higher tensile strength
steels. This permits the use of cylinder designs in which the stress in the cylinder wall is increased for a
constant wall thickness. The use of higher strength steels will therefore achieve a lower ratio of steel weight to
gas weight that reduces shipping and handling costs.
2 © ISO 2002 — All rights reserved

---------------------- Page: 7 ----------------------
ISO/TR 12391-2:2002(E)
A major concern in using higher strength steels for cylinder construction and correspondingly higher design
wall stress is the ability to maintain the same level of safety throughout the life of the cylinder. In particular,
increasing the tensile strength of the steels and increasing the stress in the wall of the cylinders could make
the cylinders less fracture resistant than cylinders made from steels with the traditionally-used lower tensile
strength levels. In order to use steels with strength levels higher than 1 100 MPa, it was determined that new
requirements were needed to assure adequate fracture resistance of the cylinders.
To develop these requirements, a Working Group on Cylinder Fracture (WG14) was formed under
ISO/TC 58/SC 3. WG 14 was assigned the task of “developing a suitable test method and specifications to
assure adequate fracture resistance for gas cylinders made from steels with tensile strengths greater than
1 100 MPa”. WG14 decided that the test method and specifications that were developed should demonstrate
that the overall “fracture resistance” of cylinders made from higher strength steels was equivalent to that of
cylinders made from lower strength steels. Fracture resistance of the cylinder is defined as the adequate
fracture initiation strength in the presence of a crack-like flaw to assure leak rather than fracture performance
of the cylinder at a specified failure pressure (usually the marked service pressure of the cylinder).
The test methods and procedures that have previously been used to evaluate the fracture performance of high
pressure cylinders have been based either on fracture mechanics tests and analysis [3] or have been based
on empirical correlations with the Charpy-V-notch (CVN) test impact energy [4]. The objectives of these tests
and analyses are to predict the fracture initiation stress (or pressure) and fracture mode (leak or unstable
fracture).
The fracture mechanics tests and analysis showed that to provide adequate fracture resistance, the cylinder
wall should be in the plane-stress fracture state and that the fracture should occur under elastic-plastic
conditions. To reliably evaluate the fracture performance of cylinders in the plane-stress fracture state requires
that an elastic-plastic fracture mechanics analysis (i.e. J , J ) be conducted. Using the fracture mechanics
Ic R
analysis approach to evaluate fracture performance may require that a complex and expensive finite-element
analysis be done for each specific type of flaw on each specific cylinder design to establish the J or J
Ic R
requirements for adequate fracture resistance. Also, the J materials property test required to evaluate the
Ic
cylinder material is expensive and time-consuming. Such costly and time-consuming tests, have not proven to
be practical for use with the high volume cylinder production.
Empirical correlations have been used to predict the fracture performance of cylinders. These empirical
correlations relate the fracture initiation stress level for specific flaw types to the Charpy-V-notch (CVN) test
impact energy. Although the Charpy-V-notch (CVN) test is useful for evaluating the quality of cylinders during
production, the Charpy-V-notch (CVN) test alone may not be a reliable means to evaluate the fracture
resistance of new designs of steel cylinders or to evaluate new alloy steels for cylinder construction.
As a result of these limitations with fracture mechanics analysis and with empirical correlations based on CVN
tests, it was concluded that an alternate approach was required to evaluate the fracture resistance of high
strength steel cylinders. It was decided that the test method that was developed should measure the total
fracture resistance of the cylinder and not just the fracture toughness. Therefore, WG 14 decided to use a
direct approach to evaluate the fracture resistance of cylinders and this led to the development of the “Flawed-
Cylinder Burst Test”.
In this test method, the fracture test is performed on an actual, full size, cylinder rather than by measuring the
fracture properties of the material alone by taking small scale test specimens from the cylinder, such as for J
Ic
tests. This test method consists of testing cylinders in which flaws of specified sizes are machined into the
external surface of the cylinders. The cylinders are pressurized until failure, and the failure pressure and
failure mode (leak or fracture) is determined. This approach is only possible because the cylinders are
required by the existing safety regulations to be produced in large, controlled groups of uniform cylinders and
therefore a single sample cylinder from the group will adequately represent the behaviour of all cylinders in the
production group.
The concept of the flawed-cylinder burst test and the development conducted under WG 14 is described in
ISO/TR 12391-1. The technical basis for the flawed-cylinder burst test is described in detail in reference [5].
In the development of the test method and acceptance criteria for the flawed-cylinder burst test, it was decided
that the fracture resistance of newer, higher-strength steel cylinders should essentially be the same as that of
the lower strength, existing cylinders because the existing cylinders have provided fracture-safe performance
© ISO 2002 — All rights reserved 3

---------------------- Page: 8 ----------------------
ISO/TR 12391-2:2002(E)
during their many years of service. Therefore, flawed-cylinder burst tests were conducted on cylinders with
strength levels covering the full range of strength levels currently being produced in the world. Tests were
conducted on cylinders made from steels ranging in tensile strength from 620 MPa to 1 400 MPa. During the
development of the flawed-cylinder burst test, several hundred flawed-cylinder burst tests were conducted by
the members of WG 14. Flawed-cylinder burst tests were conducted by 10 different companies in seven
different countries (Austria, France, Germany, Japan, Sweden, the United Kingdom and the United States).
This part of ISO/TR 12391 is limited to a summary and compilation of the results of the flawed-cylinder burst
tests that were conducted by WG 14 during the development of the flawed-cylinder burst test method. Results
of flawed-cylinder cycle burst tests that assess the fracture performance of the cylinders due to pressure
cycling were also carried out by WG 14 and are given in ISO/TR 12391-3. This part of ISO/TR 12391 is in the
form of a data base of the test results intended to be used for further analysis of the fracture performance of
steel cylinders.
5 Experimental test programme
5.1 Types of cylinder tested
Flawed-cylinder burst tests were conducted on cylinders that represented all of the currently used and
proposed new types of seamless steel cylinders. A brief description of all the cylinders that were tested is
shown in Tables 1 to 5. For this study, the cylinders were classified into material groups (designated groups A
to E) based on the actual measured tensile strength (R ) of the cylinders that were tested. The actual
m
measured tensile strength for each group of cylinders that was tested is shown in Tables 6 to 10. The general
description of the cylinders in each material group is shown below. Cylinders made from materials in groups A
to D are currently being produced and used throughout the world. Cylinders made from materials in group E,
are experimental and are not currently authorized for use.
Material group Description of cylinder Tensile strength R
m
A Cylinders made from carbon steel and which may R < 750 MPa
m
be heat treated by normalizing, normalizing and
tempering, or quenching and tempering
B Cylinders made from alloy steel (Cr-Mo steels) heat
750 MPa u R u 950 MPa
m
treated by quenching and tempering; these
cylinders may generally be used for all gases
C Cylinders made from alloy steel (Cr-Mo steels) heat 950 MPa u R u 1 080 MPa
m
treated by quenching and tempering; these
cylinders are restricted to use with non-corrosive
gases and are made in accordance with ISO 9809-1
D Cylinders made from alloy steel (Cr-Mo steels) heat 1 080 MPa u R u 1 210 MPa
m
treated by quenching and tempering; high strength
and high toughness steel cylinders restricted to use
with non corrosive gases and made in accordance
with ISO 9809-2
E Experimental cylinders; extra high strength; not
R > 1 210 MPa
m
currently authorized for use

Within each main material group (A to E) material subgroups are designated, e.g., material subgroup A-1, A-2.
All the cylinders within a given subgroup were made to the same specification, of the same size (diameter,
thickness, and volume), the same material, the same specified tensile strength range, the same designated
service pressure and test pressure and were made by the same manufacturing process. The cylinders in a
specific material subgroup (e.g. subgroup B-2) may be of a different alloy, size, design specification or
4 © ISO 2002 — All rights reserved

---------------------- Page: 9 ----------------------
ISO/TR 12391-2:2002(E)
manufacturing process than cylinders in a different materials subgroup (for example B-3) in the same main
material group (e.g. group B). However, the actual measured tensile strength for all cylinders in a material
group will be in the same range (e.g., 750 MPa to 950 MPa for all cylinders in group B).
In Tables 1 to 5, it should be noted that the code numbers for some material subgroups (e.g. B-1, C-1 and
C-2) are missing. The cylinders in these missing material subgroups were tested using the flawed-cylinder
burst test with cyclical pressurization and the results are given in ISO/TR 12391-3.
In Tables 1 to 5, each flawed-cylinder burst test is assigned a number in sequence, as shown in the first
column, for purposes of tracking each test. The same number is then used to identify the cylinders in the
tables for the results of the mechanical properties tests (Tables 6 to 10) and in the tables for the results of the
burst test (Tables 12 to 16). In addition, each individual cylinder tested is assigned a number, such as A-1-1,
as shown in the second column of the tables.
The specified tensile strength range given in Tables 1 to 5 is the range of “guaranteed” minimum, R , and
g, min
maximum, R , tensile strength designated by the cylinder manufacturer or the cylinder specification used
g, max
for the design of the cylinder. These values are used to calculate the cylinder wall thickness when designing
the cylinder. These are specified values rather than actual measured values of the tensile strength, R . In a
m
few cases, the manufacturer did not provide a specified minimum or maximum tensile strength values.
The information required to calculate the wall thickness of the cylinder, the test pressure of the cylinder and
the service pressure of the cylinder is listed in Tables 1 to 5. This information includes the outside diameter of
the cylinder, D, and the particular national or international design specification used by the manufacturer to
design the cylinder. These specifications are used to calculate the stress in the cylinder wall, the minimum
design wall thickness of the cylinder, t , the maximum design test pressure, P , and the maximum design
d h
service pressure, P . Each of the national or international cylinder specifications has a different formula for
s
calculating the stress in the wall of the cylinder and therefore the design wall thickness for a specified cylinder
diameter and service pressure. In some cases the cylinders tested were not designed to an existing design
specification so these cylinders are designated as experimental cylinders.
The other items listed in Tables 1 to 5, for information purposes only, are the type of manufacturing process
used to make the cylinder, the cylinder volume (in litres) and the specific material used, when given. This
information is shown only to better identify the cylinders that were tested and is not used for any analysis of
the test results.
In Tables 1 to 5, the results show that in some cases the same cylinder was tested several times. This was
achieved by welding the cylinder shut after it had leaked and re-testing it until it failed by fracturing. In this
case the cylinder numbering sequence is shown repeatedly as the same cylinder number, e.g. as A-1-1, but
the burst test number is shown sequentially as number 1, 2, 3 and 4. In other cases, a different cylinder was
used for each burst test in the material subgroup series and each cylinder was tested only once. In these
cases, each cylinder will have the same material subgroup number but will have a different cylinder number.
An example of this is shown for material subgroup B-3 where the cylinders tested are numbered as B-3-1,
B-3-2, etc.
In Tables 1 to 5, there are a few cases, such as material subgroup B-2, where the specified tensile strength
range (e.g., 1 069 MPa to 1 207 MPa) does not agree with the tensile strength range for that particular material
groups (e.g., 750 MPa to 950 MPa). In these cases, the cylinders were manufactured to a particular strength
range (e.g., 1 069 MPa to 1 207 MPa) but were then re-tempered to change their actual strength for use in the
studies reported here. For these studies, the test results for these cylinders were put into the material group
represented by the actual measured tensile strength range and not the tensile strength range represented by
the specified range.
It should be noted that in a few cases, the actual measured tensile strength (R ) for one or more cylinders in a
m
particular material subgroup is slightly outside the designated range for the tensile strength of the particular
material subgroup in which the cylinder is included. However, the measured tensile strength of the rest of the
cylinders from that material subgroup that were tested are within the appropriate tensile strength range for that
material subgroup. Examples of this occur in material subgroups A-1, B-9, C-5, D-9 and D-10.
© ISO 2002 — All rights reserved 5

---------------------- Page: 10 ----------------------
ISO/TR 12391-2:2002(E)
5.2 Material properties tests
Conventional mechanical properties tests, such as tensile tests and Charpy-V-notch tests, were conducted on
each set of cylinders on which flawed-cylinder burst tests were performed. The results of these tests are
shown in Tables 6 to 10 for each group of materials.
The tensile test results shown in Tables 6 to 10 are the actual measured yield strength, R , the actual
ea
measured tensile strength, R , and the total elongation, A. These material properties are required to be
m
measured by all of the existing national or international cylinder design specifications. The actual measured
tensile strength, R value is used to determine that the cylinder meets the specification to which
...