ISO/TR 12391-3:2002
(Main)Gas cylinders — Refillable seamless steel — Performance tests — Part 3: Fracture performance tests — Cyclical burst tests
Gas cylinders — Refillable seamless steel — Performance tests — Part 3: Fracture performance tests — Cyclical burst tests
ISO/TR 12391-3 applies to seamless refillable cylinders of all sizes from 0,5 l up to and including 150 l water capacity produced of steel with tensile strength greater than 1 100 MPa. It can also be applied to cylinders produced from steels used at lower tensile strengths. In particular, it provides the technical rationale and background to guide future alterations of existing ISO standards or for developing advanced design standards. ISO/TR 12391-3 is a summary and compilation of the test results obtained during the development of the Flawed-cylinder Cyclical-burst Test. The test is an alternate test method to the flawed-cylinder burst test with monotonic pressurization and is used to evaluate the fracture performance of steel cylinders which are used to transport high-pressure compressed gases. In ISO/TR 12391-3 test results are reported for more than one hundred flawed-cylinder cyclical burst tests that were conducted on seamless steel cylinders that ranged in tensile strength from 750 MPa to 1 210 MPa. The test method is intended to be used both for the selection of materials and to establish design parameters in the development of new cylinders 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 3: Essais de mode de rupture — Essais de rupture cyclique
General Information
Relations
Standards Content (Sample)
TECHNICAL ISO/TR
REPORT 12391-3
First edition
2002-12-15
Gas cylinders — Refillable seamless
steel — Performance tests —
Part 3:
Fracture performance tests — Cyclical
burst tests
Bouteilles à gaz — Rechargeables en acier sans soudure — Essais de
performance —
Partie 3: Essais de mode de rupture — Essais de rupture cyclique
Reference number
©
ISO 2002
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 copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2002 — All rights reserved
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. 5
5.3 Description of the flawed-cylinder cyclical burst test. 6
6 Flawed-cylinder cyclical burst test results. 8
6.1 Flawed-cylinder burst test procedure. 8
6.2 Flawed-cylinder cyclical burst test results for group B materials. 9
6.3 Flawed-cylinder cyclical burst test results for group C materials. 10
6.4 Flawed-cylinder cyclical burst test results for group D materials. 11
7 Discussion . 13
7.1 Background . 13
7.2 ISO 9809-2 flawed-cylinder cyclical burst test procedures and acceptance criteria. 13
7.3 Comparison of the flawed-cylinder cyclical burst test with the flawed-cylinder burst test
with monotonic pressurization to evaluate fracture performance. 14
8 Summary and conclusions . 15
Bibliography . 47
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 part of ISO/TR 12391 may be the subject
of patent rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO/TR 12391-3 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
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 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.
TECHNICAL REPORT ISO/TR 12391-3:2002(E)
Gas cylinders — Refillable seamless steel — Performance
tests —
Part 3:
Fracture performance tests — Cyclical burst tests
1 Scope
This part of ISO/TR 12391 applies to seamless refillable cylinders of all sizes from 0,5 l up to and including
150 l water capacity produced of steel with tensile strength (R ) greater than 1 100 MPa.
m
It can also be applied to cylinders produced from steels used at lower tensile strengths. In particular, it
provides the technical rationale and background to guide future alterations of existing ISO standards or for
developing advanced design standards.
This part of ISO/TR 12391 is a summary and compilation of the test results obtained during the development
of the “flawed-cylinder cyclical burst test”. The test is an alternate test method to the flawed-cylinder burst test
with monotonic pressurization and is used to evaluate the fracture performance of steel cylinders which are
used to transport high-pressure compressed gases.
The concept and development of the flawed-cylinder cyclical burst test is described in ISO/TR 12391-1. The
details of the test method and the criteria for acceptable fracture performance of steel cylinders are given in
9.2.5.3.2 of ISO 9809-2:2000. In this part of ISO/TR 12391, test results are reported for more than one
hundred flawed-cylinder cyclical burst tests that were conducted on seamless steel cylinders that ranged in
tensile strength from 750 MPa to 1 210 MPa. The test method is intended to be used both for the selection of
materials and to establish design parameters in the development of new cylinders 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 12391-2, Gas cylinders — Refillable seamless steel — Performance tests — Part 2: Fracture
performance tests — Monotonic burst tests
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 guaranteed minimum yield strength;
e
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, etc.) are stored and
transported in portable steel cylinders. These cylinders are designed, manufactured, and maintained in
accordance with ISO 9809-1 and ISO 9809-2. The cylinders are constructed from specified alloy steels that
[1]
are generally modified versions of steel alloys such as AISI 4130 or 34 Cr Mo 4 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 by 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 from 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, which are modifications of the AISI 4130 and AISI 4140 steels, 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
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 out of 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 (WG 14) 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". WG 14 decided that the test method and specifications that were developed should demonstrate
that the overall “fracture resistance” of cylinders made out of 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 or have been based on
empirical correlations with the Charpy-V-notch (CVN) test impact energy [4]. The objectives of these test
methods and procedures 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 steel alloys 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” and the “Flawed-cylinder cyclical burst test”.
In these test methods, 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 tests. The proposed test methods consist of testing cylinders in which flaws of specified sizes are
Ic
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 flawed-cylinder cyclical burst test and the development
conducted under WG 14 are described in ISO/TR 12391-1. The results of the flawed-cylinder burst test
conducted with monotonic pressurization are described in ISO/TR 12391-2.
In the development of the test method and acceptance criteria for the flawed-cylinder burst test and the
flawed-cylinder cyclical burst test, it was decided that the fracture performance of newer, higher-strength steel
cylinders should be essentially the same as that of the lower strength, existing cylinders because the existing
cylinders have provided fracture-safe performance during their many years of service. Therefore, flawed-
cylinder burst tests and flawed-cylinder cyclical burst tests were conducted on cylinders with strength levels
covering the full range of strength levels currently being produced in the world. Flawed-cylinder burst tests
were conducted on cylinders made from steels ranging in tensile strength from 620 MPa to 1 400 MPa and
flawed-cylinder cyclical burst tests were conducted on cylinders made from steels ranging in tensile strength
from 750 MPa to 1 210 MPa. Ten different companies in seven different countries (Austria, France, Germany,
Japan, Sweden, the United Kingdom and the United States) conducted flawed-cylinder burst tests and flawed-
cylinder cyclical burst tests.
This part of ISO/TR 12391 is limited to a summary and compilation of the results of the flawed-cylinder cyclical
burst tests that were conducted by WG 14 during the development of the test method for the flawed-cylinder
cyclical burst tests. This part of ISO/TR 12391 is in the form of a database of the test results that is 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 cyclical 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 3. For this study, the cylinders were classified into material groups (designated groups B
to D) 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 4 to 6. The general
description of the cylinders in each material group is shown below. Cylinders made from materials in groups B
to D are currently being produced and used throughout the world.
NOTE Material groups designated A and E were only used for cylinders tested using the flawed-cylinder burst test
with monotonic pressurization and the results are reported in ISO/TR 12391-2.
Material group Description of cylinder Tensile strength R
m
B Cylinders made from alloy steel (Cr-Mo steels) heat 750 MPa < 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
gasesand 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 gasesand and made in
accordance with ISO 9809-2
Within each main material group (B to D) material subgroups are designated, e.g., material subgroups B-1
and B-10. 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-10) may be of a different alloy, size, design
specification or manufacturing process than cylinders in a different materials subgroup (e.g. B-1) 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 3, it should be noted that the code numbers for some material subgroups (e.g. B-2, C-3 and
C-4) are missing. The cylinders in these missing material subgroups were tested using the flawed-cylinder
4 © ISO 2002 — All rights reserved
cyclical burst test, with monotonic pressurization and were not tested using the flawed-cylinder cyclical burst
test procedures. The results of the tests on cylinders made from missing material subgroups using the flawed-
cylinder cyclical burst test, with monotonic pressurization are reported in ISO/TR 12391-2.
In Tables 1 to 3, each flawed-cylinder cyclical 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 property tests (Tables 4 to 6) and in the tables for the results of the
flawed-cylinder cyclical burst test (Tables 7 to 9). In addition, each individual cylinder tested is assigned a
code, such as CB-B-1-1 as shown in the second column of the tables; e.g. this code shows that the test is a
cyclical burst test (CB), for material subgroup B-1, and is cylinder number 1 in this material subgroup.
The specified tensile strength range shown in Tables 1 to 3 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 are shown in Tables 1 to 3. This information includes the outside diameter
or the cylinder, D, and the particular national or international design specification (when provided by the
manufacturer) 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
d
pressure, P and the maximum design service pressure, P . Each of the national or international cylinder
h s
specifications may have a different formula for 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 shown in Tables 1 to 3, for information purposes only, are:
the type of manufacturing process used to make the cylinder;
the cylinder volume (in litres);
the specific material used (when provided by the manufacturer).
This information is only shown in order to better identify the cylinders that were tested and is not used for any
analysis of the test results.
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 the same material subgroup that were tested is within the appropriate tensile strength range for
that material subgroup.
5.2 Material properties tests
Conventional mechanical properties tests, such as tensile tests conducted in accordance with ISO 6892 and
Charpy-V-notch tests conducted in accordance with ISO 148, were carried out on each set of cylinders on
which flawed-cylinder cyclical burst tests were performed. The results of these tests are shown in Tables 4 to
6 for each group of materials.
The tensile test results shown in Tables 4 to 6 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 value is used to determine that the cylinder meets the specification to which it is manufactured
and is used in this test programme to classify the cylinder in the appropriate material group tested. The actual
measured yield strength is used to determine that the cylinder meets the requirement for the yield strength to
tensile strength ratio when this ratio is a part of the specification. The actual measured tensile strength value
may also be used for additional analysis of the cylinder design parameters permitted in some of the
specifications. The total elongation is used to determine that the requirement for minimum elongation is met
when that is part of the specification to which the cylinder is manufactured. The elongation value is not used
for any calculations in the design of cylinder.
For cylinders manufactured in the United States (such as those designated as DOT type 3A or 3AA) the
tensile tests used to measure the properties of the cylinders were of the type specified by the CFR Title 49
[2]
Part 178 . These test specimens have a fixed gauge length of 50 mm, a fixed width of 38 mm and a
thickness equal to the actual wall thickness of the finished cylinder from which the specimens were taken. For
other cylinders, the tensile tests of the type specified by ISO 6892 were used. The ISO test specimens have a
gauge length of 5,65 × the square root of the cross section of the specimen, a width of 4 × the specimen
thickness and a thickness equal to the actual wall thickness of the finished cylinder from which the specimens
were taken. The ultimate tensile strength and the yield strength values should be essentially the same when
measured with either the DOT or the ISO type of specimen. The measured elongation values will be different
depending on the specific type of tensile specimen used.
[3]
The Charpy-V-notch tests were conducted in accordance with test method ASTM E23-02 for cylinders
manufactured in the United States. For other cylinders, the Charpy-V-notch tests were conducted in
accordance with the test method described in ISO 148. The Charpy-V-notch impact test energy values should
be essentially the same when measured with either the ASTM or the ISO test method. The Charpy-V-notch
test specimens had cross sectional dimensions of either 10 mm deep by 5 mm thick or 10 mm deep by 4 mm
thick depending on the available wall thickness of the cylinder and the orientation of the Charpy-V-notch test
specimen. The exact dimensions of each Charpy-V-notch test specimen used are shown in Tables 4 to 6. The
Charpy-V-notch tests were conducted either at ambient temperature (20 °C ) or at low temperature (− 50 °C ),
as shown in Tables 4 to 6.
The Charpy-V-notch test specimens were either oriented with the longitudinal axis of the specimen parallel to
the longitudinal axis of the cylinder (designated longitudinal specimens) or with the longitudinal axis of the
specimen perpendicular to the longitudinal axis of the cylinder (designated transverse specimens). As shown
in Tables 4 to 6, not all combinations of test temperatures and specimen orientation were used on each
cylinder that was tested. The total energy absorbed in breaking the Charpy-V-notch test specimens was
measured in Joules (J). All Charpy-V-notch test results are reported in J/cm , where the total energy absorbed
is divided by the area of the specimen ligament below the specimen notch.
In the specifications for certain cylinder designs, particularly for material groups C and D cylinders, minimum
Charpy-V-notch energy levels are required. The Charpy-V-notch tests were conducted on all cylinders to
determine that these requirements were met. The Charpy-V-notch energy test results are not used to evaluate
the results of the flawed-cylinder cyclical burst test. However, the Charpy-V-notch energy test results are
reported here because these results may be used to evaluate the fracture performance of the cylinders using
alternate analysis procedures to the flawed-cylinder cyclical burst test.
For certain material subgroups on which flawed-cylinder cyclical burst tests were conducted, mechanical
properties test specimens were taken from each cylinder in the material subgroup after the burst test was
completed. In this case, the test results are shown in the tables of results for each of the individual cylinders.
For some material subgroups on which flawed-cylinder cyclical burst tests were conducted, mechanical
properties test specimens were taken only from selected cylinders in that particular material subgroup after
the burst test was completed. In these cases, results are shown in the tables of results for the cylinders for
which mechanical property tests were conducted and blank spaces are shown for the other cylinders on which
flawed-cylinder cyclical burst tests were conducted but mechanical properties tests were not conducted.
Because the cylinders in a particular material subgroup are all of the same type and from the same production
batch, the mechanical properties test results for the cylinders that were tested are considered to adequately
represent the properties of all cylinders in that material subgroup.
5.3 Description of the flawed-cylinder cyclical burst test
The flawed-cylinder cyclical burst test is used to evaluate the overall fracture performance of the entire
cylinder and not just the “fracture toughness” of the material as determined with conventional fracture
toughness test specimens. The flawed-cylinder cyclical burst test is intended to be both a “design qualification
6 © ISO 2002 — All rights reserved
approval test” and a “production lot test”. The full details of the test and the criteria for acceptable fracture
performance of steel cylinders are given in 9.2.5 of ISO 9809-2:2000.
In the flawed-cylinder cyclical burst test, the fracture performance of the cylinder is evaluated by cyclically
pressurizing a cylinder with a designated type (shape and sharpness) and size (length and depth) of surface
flaw until failure occurs. Failure occurs either by leaking or by fracturing.
The cylinder to be tested has a flaw machined into the exterior surface of the cylinder wall. The flaw is
machined in the location of probable maximum stress under pressurized loading, i.e. a longitudinal surface
flaw at mid-length and at the thinnest place in the cylinder wall. To make the tests adequately uniform and
reproducible, a surface flaw with a deep enough standard geometry is required. The geometry of the standard
flaw is shown in Figure 1. A standard Charpy-V-notch milling cutter is used to machine the flaw to the
designated length and depth. The milling cutter is required to meet the following specification:
thickness of the cutter = 12,5 mm ± 0,2 mm;
angle of the cutter = 45° ± 1°;
tip radius u 0,2 mm;
for cylinders u 140 mm in diameter, cutter diameter = 50,0 mm ± 0,5 mm;
for cylinders > 140 mm in diameter, cutter diameter = 60 mm to 80 mm.
This machined flaw is a surface flaw of the type shown in Figure 1. The flaw size is specified as:
0,5
flaw length, l = 1,6 (D × t )
o d
NOTE For the size of cylinders tested here, this flaw length is approximately l = 10 × t .
o d
flaw depth, d > 0,60 × t
d
Pressurization is carried out hydrostatically. During the test, each cylinder is filled with water at room
temperature and the pressure is cycled until the cylinder fails. The maximum cycling pressure is the “adjusted
design service pressure” = P × (t /t ). This pressure is also the designated failure pressure, P , of the cylinder
s a d f
in the test. The minimum cycling pressure is 10 % of the maximum pressure, but not more than 30 bar. The
maximum cycling frequency is 15 cycles/min.
The test is continued until failure occurs. The number of cycles to failure is recorded and the failure mode
(either leak or fracture) is recorded. For this test the definition of fracture is “an extension of at least 10 % in
the length of the machined flaw in the longitudinal direction”. The failure pressure and failure mode, either leak
or fracture, are reported as the test results. The failure pressure is the maximum cyclical pressure,
P = P × (t /t ). During the flawed-cylinder cyclical burst test, a single test cylinder may be sufficient if the
f s a d
failure occurs by leaking, and shows that the cylinder has adequate fracture resistance.
The overall fracture performance of the cylinder is determined with the flawed-cylinder cyclical burst test by
empirically determining the “leak-fracture boundary” for the specified maximum cycling pressure. This requires
that a series of cylinders with different flaw lengths be tested. The “leak-fracture boundary” for a specified
maximum cycling pressure is defined as the average of the longest flaw length at which a leak occurs and the
shortest flaw length at which a fracture occurs.
During the development of the flawed-cylinder cyclical burst test, some tests were conducted on series of
cylinders with a range of flaw lengths to define the “leak-fracture boundary” for each particular type of cylinder
and material. This was done to evaluate the overall fracture performance of the cylinder type. An example of
such test results is shown in Figure 2. It is expected that this procedure to determine the full “leak-fracture
boundary” over a range of flaw lengths will be used only for the “design qualification” evaluation of new
cylinders (i.e. for new materials and production processes) to demonstrate that the cylinder type has adequate
fracture resistance.
Once the full fracture performance is determined for a particular cylinder type from the flawed-cylinder cyclical
burst tests conducted during the “design qualification” procedure, the testing procedure used to evaluate
cylinders during large scale production can be simplified and made much more efficient. For production testing,
a single cylinder with the specified flaw length is tested at the designated maximum cycling pressure and if the
failure mode is a ”leak”, the cylinder satisfies the criteria for adequate fracture resistance given in ISO 9809-2.
During the development of the flawed-cylinder cyclical burst test, it was decided that the acceptable level of
fracture resistance for cylinders of any strength level should be equivalent to the fracture resistance of existing
cylinders that have been used for extended periods of time. Therefore, flawed-cylinder cyclical burst tests
were conducted on cylinders with tensile strength levels ranging from about 750 MPa to 1 210 MPa. The
existing cylinders (with tensile strengths levels less than 950 MPa) have provided fracture-safe performance
during many years of service. From these results, it was determined that to have fracture resistance
equivalent to the fracture resistance of existing cylinders, new higher strength steel cylinders should have a
0,5
leak-fracture boundary of P /P greater than 1,0 when the designated flaw length was, l = 1,6 (D × t ) .
f s o d
6 Flawed-cylinder cyclical burst test results
6.1 Flawed-cylinder burst test procedure
The results of all of the flawed-cylinder cyclical burst tests that were conducted are shown in Tables 7 to 9.
For each cylinder tested, the crack length, l , in terms of a multiple of the design minimum cylinder wall
o
thickness, t , is given as l = n × t (e.g. l = 10 t ). This term is used as a common reference to compare
d o d o d
cylinders with different wall thicknesses. The flaw depth, d, at the start of the cycling testing is given as a
percentage of the design minimum cylinder wall thickness t (e.g. 100 × d/t = 80 %).
d d
For a specified flaw length and cycling pressure, the number of cycles at which the cylinder fails depends on
the initial depth of the flaw and the thickness of the remaining ligament of metal below the flaw. Failure of the
cylinder occurs when the original flaw depth is increased by fatigue due to the pressure cycling so that the
ligament of metal below the flaw breaks. Once the ligament of metal below the flaw fails, the cylinder will
either leak or fracture depending on whether the combination of stress and flaw length is below or above the
critical level for fracture to occur.
The actual cylinder wall thickness, t , at the location of the machined flaw is measured after the test. The
a
actual cylinder wall thickness at any location in the cylinder should be greater than the design minimum
cylinder wall thickness, t . The actual measured cylinder wall thickness is included in the data in order to
d
permit additional analysis of the results using this cylinder wall thickness instead of the nominal cylinder wall
thickness that is given by the design minimum cylinder wall thickness. The pressure at the time that the
cylinder fails, either by leaking or by fracturing, is given as P measured in bar. The failure mode, either leak or
f
fracture, is reported.
The ratio of the failure pressure, P , to the marked service pressure of the cylinder, P , is given as P /P . The
f s f s
marked service pressure (bar) is the maximum pressure to which the cylinder may be filled when in service
and is specified by the cylinder manufacturer. It should be noted that the marked service pressure for the
cylinders of the same size and tensile strength would be slightly different depending on the design
specification used by the manufacturer. The cylinders were designed and the marked service pressure was
specified according to the design specification used in the country of manufacture. The use of the parameter
P /P permits the leak-fracture boundary to be defined in terms of the marked service pressure of the cylinder.
f s
This allows a comparison of cylinders of different sizes (diameters and wall thickness) to be made on a
common basis.
ISO 9809-2 requires that the measured ratio of the failure pressure to the service pressure, P /P , be adjusted
f s
to account for the local thickness of the cylinder wall at the location of the flaw. This adjustment was made to
the measured values of the P /P ratio for all of the flawed-cylinder cyclical burst tests conducted in this study.
f s
The adjusted ratio of the failure pressure to the service pressure, P /P , is shown as the last column in
f, adjusted s
Tables 7 to 9.
For completeness of the database, the results of all tests that were conducted are shown in Tables 7 to 9. It
should be noted that a range of test parameters (such as flaw length, flaw depth and maximum cycling
8 © ISO 2002 — All rights reserved
pressure) were used in the tests conducted here as part of the development of the flawed-cylinder cyclical
burst procedure. Therefore, many of the tests were not conducted using the exact flaw length, flaw depth, and
maximum cycling pressure specified in ISO 9809-2 and shown above.
For some of the material subgroups, cylinders with a range of flaw lengths were tested to define the full leak-
fracture boundary over a range of flaw lengths. The results of flawed-cylinder cyclical burst tests for these
material subgroups are shown in Tables 7 to 9 and are plotted in the Figures 3 to 7. To define the leak-
fracture boundary, the shortest flaw length for which a failure occurred by fracture and the longest flaw length
for which failure occurred by leaking are plotted. An estimate of the leak-fracture boundary is shown in
Figures 3 to 7 for each material subgroup. Data for failures over a range of flaw lengths is available for
material subgroups B-1, C-5, D-1, D-7 and D-12.
For some of the other material subgroups, all the flawed-cylinder cyclical burst tests were conducted at a
single defined flaw length and both leak results and fracture results were obtained. For these tests the leak-
fracture boundary can be defined only for the single specified flaw length. These results are not plotted but the
test results are summarized and the estimated leak-frac
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.
Loading comments...