ITU Lifetime
ITU Lifetime
ITU-T Series G
TELECOMMUNICATION Supplement 59
STANDARDIZATION SECTOR
OF ITU (09/2016)
Summary
Supplement 59 to ITU-T G-series Recommendations provides guidance regarding the long term
reliability of cabled optical fibres. This Supplement uses currently accepted models combined with
current experience to describe items that can impact the performance of an optical fibre over time. The
document describes ''optical reliability'' for fibres, ''mechanical reliability'' for fibres and describes how
optical cables impact these properties.
History
Edition Recommendation Approval Study Group Unique ID*
1.0 ITU-T G Suppl. 59 2016-09-30 15 11.1002/1000/13123
* To access the Recommendation, type the URL http://handle.itu.int/ in the address field of your web
browser, followed by the Recommendation's unique ID. For example, http://handle.itu.int/11.1002/1000/1
1830-en.
NOTE
In this publication, the expression "Administration" is used for conciseness to indicate both a
telecommunication administration and a recognized operating agency.
Compliance with this publication is voluntary. However, the publication may contain certain mandatory
provisions (to ensure, e.g., interoperability or applicability) and compliance with the publication is achieved
when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as
"must" and the negative equivalents are used to express requirements. The use of such words does not suggest
that compliance with the publication is required of any party.
ITU 2016
All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior
written permission of ITU.
1 Scope
The scope of this Supplement is to give the end user guidance on the long-term performance of optical
fibre and cables. Though it is difficult to address all situations and to guarantee long-term
performances, this supplement helps understand general long-term behaviour of optical fibres and
cables and provides guidelines to help the end users minimize the number of mechanical and optical
failures during the expected lifetime of the fibre and cable.
2 References
[ITU-T G.651.1] Recommendation ITU-T G.651.1 (2007), Characteristics of a 50/125 µm
multimode graded index optical fibre cable for the optical access network.
[ITU-T G.652] Recommendation ITU-T G.652 (2016), Characteristics of a single-mode
optical fibre and cable.
[ITU-T G.653] Recommendation ITU-T G.653 (2010), Characteristics of a dispersion-shifted,
single-mode optical fibre and cable.
[ITU-T G.654] Recommendation ITU-T G.654 (2016), Characteristics of a cut-off shifted
single-mode optical fibre and cable.
[ITU-T G.655] Recommendation ITU-T G.655 (2009), Characteristics of a non-zero
dispersion-shifted single-mode optical fibre and cable.
[ITU-T G.656] Recommendation ITU-T G.656 (2010), Characteristics of a fibre and cable
with non-zero dispersion for wideband optical transport.
[ITU-T G.657] Recommendation ITU-T G.657 (2016), Characteristics of a bending-loss
insensitive single-mode optical fibre and cable for the access network.
[ITU-T G-Sup.40] Supplement ITU-T G.Sup40 (2010), Optical fibre and cable
Recommendations and standards guideline.
[ITU-T L.126] Recommendation ITU-T L.126/L.27 (1996), Method for estimating the
concentration of hydrogen in optical fibre cables.
3 Definitions
None.
5 Conventions
None.
7 Optical reliability
In the 30 years of commercial optical cable deployment, several mechanisms have been identified
that deteriorate the optical performance of glass fibres. These mechanisms can be considered as
starting points in identifying processes that need to be accounted for to assure the optical
performances of manufactured and cabled fibres prescribed in ITU-T G.65x series Recommendations.
This clause focuses on three known mechanisms that have an impact on optical reliability:
– Bending loss (microbending and macrobending)
– Chemical attacks (hydrogen being the most common contributor)
– Radiation sensitivity
Many of the tests performed on optical fibre and cable found in IEC and ITU-T documents were
developed to help assure long term reliability of optical cables. All the tests focused on optical
reliability and monitored the fibre or cabled attenuation during the course of the test to see if any
change occurs. If the change is under the recommended threshold the optical cable is expected to
operate in the field for a reasonable period of time (typically 20-30 years) without significant
degradation in the transmission properties.
7.2 Hydrogen
Hydrogen is a small molecule that can easily diffuse into the glass structure. Hydrogen can react with
glass to create irreversible absorption peaks (Type 1) that add to the optical loss or interstitial
molecular hydrogen that can be present in the glass adding to the attenuation (Type 2) [b-Lemaire].
Both mechanisms are of concern in deployed optical cables as they can add to the overall attenuation
of the optical cable.
Interstitial hydrogen (Type 2) is only a problem where optical cable is subject to high partial pressures
of hydrogen. One application where this type of hydrogen is of concern is underwater cables. Optical
cables used in deep water, over 100 meters below the surface of the water, are put in hermetic
packages, most commonly stainless steel tubes, to prevent Type 2 absorption from occurring. Type 2
absorption in optical cables deployed in shallow water (less than 100 m in depth) increases due to
hydrogen, and the impact increases with depth. The impact of Type 2 hydrogen is reversible. Once
the cable is removed from the water, the hydrogen diffuses out of the cable and attenuation returns to
normal.
Hydrogen can react with defects in the glass matrix. These Type 1 defects are most apparent in
ITU-T G.652.B fibres. These defects form as hydrogen reacts with the glass resulting in OH groups
which – if created in sufficient quantity in the light guiding portion of the optical fibre – add to the
attenuation of the optical fibre across the full useable spectrum. In sufficient quantity these defects
can cause the optical fibres to become unusable.
In 2001 ITU-T introduced full spectrum ITU-T G.652 fibre. These ITU-T G.652 fibres differed from
their predecessors in that the glass that is transmitting the light is stabilized so that under typical
deployment conditions Type 1 hydrogen defects do not form in sufficient quantity to adversely impact
the transmission properties of the optical fibre. Today most optical fibres meet this specification
providing more reliable long-term optical performance. If ITU-T G.652 fibres are exposed to
hydrogen concentrations significantly higher than what occurs in ambient conditions these defects
Figure 1 – Failure probability for over 100 km of fibre tested at 10 m gauge lengths
[b-Mazzarese]
The estimates for fibre reliability are based on a flaw distribution that has two regions: a strong
''intrinsic strength'' and a weaker ''extrinsic strength''. The intrinsic region is characteristic of short
lengths of fibres and can be determined by testing small amounts of fibre. The intrinsic strength has
median value typically greater than 3 GPa (about 450 kpsi). The extrinsic strength distribution
characterizes larger flaws that occur at a much lower frequency and that are typically found only by
testing long lengths (typically 10-20 metre) of fibre, as in the testing represented above. The
maximum allowable flaw size in a population of optical fibre is fixed by proof-testing the fibres,
removing the weakest flaws, and this action is normally applied to 100% of commercial fibre by
manufacturers at the end of the drawing and coating phases. Currently optical fibres are proof tested
to 0.69 GPa proof stress (100 kpsi), which screens flaws that have a depth of 0.5 micron or larger.
The variables used in the equation and conservative values often used in calculations are given below:
Parameters Description/example
tf is the time to failure (lifetime) calculated
tp is the proof test time see clause 8.5
σ𝑝 is the proof stress 0.69 GPa (100 kpsi )
σ𝑎 is the applied load see clause 10
F is the failure probability See Appendix II
Np is the proof test break rate 1 break per 100 km
L is the length under tension determined by application
n is the stress corrosion parameter See clause 8.4
md is the Weibull m from the fibre dynamic tensile See clause 8.4
strength
8.4 Static and dynamic stress corrosion parameter and impact on reliability
The three fibre parameters that have the greatest impact on the lifetime predicted in equation 8-1 are
the proof-stress (σ𝑝 ) , the stress corrosion parameter (n), and the Weibull slope (m). As explained in
clause 8.1, the proof-stress is set by the manufacturers and is used as a screen to eliminate the largest
flaws. The stress corrosion parameter must be measured and that measurement can be challenging.
This parameter represents the susceptibility of the fibre glass to withstand stress corrosion and
subsequent increase of flaws due to the water effect.
Though the stress corrosion factor is shown as a constant in equation 8-1, the value obtained is
dependent on the conditions under which the parameter is measured. The limits of this parameter are
the static stress corrosion parameter (ns), which tends to provide higher values of n, and the dynamic
stress corrosion parameter (nd), which tends to provide lower values. What further complicates this
analysis is that it has been shown that the value of n is dependent on the rate of strain applied to the
fibre. In the work of COST 218 [b-IEC 60793-1-33] it was shown that the value of n varied from a
value of ~17 for the fastest tests to ~40 for very slow tests. The power law reliability model does not
specify which value of n should be used in the equation, nevertheless lower values means a more
conservative approach (worst case).
Larger values of n result in longer time until failure in equation 8-1. Alternatively, a larger value of
n implies that a larger permanent strain can be allowed in the fibre by the cable design. Thus
conservative engineering principles suggest that a fibre supplier shows that their fibre demonstrates
a value for the dynamic stress corrosion parameter greater than 18 [b-IEC 60793-2-50] and n~20 is
typically used in predicting typical lifetimes. Higher values than 20 will not translate into
demonstrable enhanced fatigue resistance.
If the cable manufacturer or end-user should wish to design for more aggressive loading of the optical
fibres, or for longer anticipated lifetime in service (e.g., up to 100 years), the value of n for the fibre
population should be characterized carefully and by the appropriate methodology to obtain the most
accurate determination of the stress corrosion parameter. For example, [b-IEC 60793-1-33] indicates
that any silica fibre mechanical test should determine the fracture stress and stress corrosion
properties under conditions that model the practical applications as closely as possible. In such a
situation, the static stress corrosion parameter approach to characterizing the n-value can be useful,
representing stress situation most similar to installed fibre. Since it has been found that large, extrinsic
region flaws (the danger to long lengths of fibre under stress) exhibit a stress corrosion factor larger
than found in the intrinsic strength flaws [b-Griffioen2], [b-Craig], [b-Glaesemann2], [b-Breuls],
[b-Yuce] static stress corrosion parameter is arguably both less conservative and more accurate.
The Weibull slope (m) corresponds to the slope of failure probability in the extrinsic region, i.e., the
slope in Region II in Figure 8-1. Accurate characterisation of this parameter requires measuring the
mechanical strength of many kilometres of fibre, using a long gauge length test and determining the
m value from information such as what is shown in Figure 8-1. This low strength tail often determines
the long term mechanical reliability of deployed fibre cables.
9 Strain due to bends tension, to axial tension or a combination of the two mechanisms
Flaws grow in an optical fibre when it is under stress. This stress can be the result of axial tension, a
small bend in the fibre or a combination of the two mechanisms. In determining the stresses imposed
on optical fibres it is essential to understand the deployment conditions, installation practices as well
as the cable design.
NOTE – The reliability of a small bending radius is an ongoing topic under ITU-T and IEC study.
I.1 Introduction
Fibres under installation at a reduced bending radius including multi-dwelling units (MDUs) and
closures may impose concerns with respect to fibre lifetime expectation. Important parameters that
determine the expected lifetime are the extrinsic and intrinsic strength in a fibre. The required values
of these parameters have to be offset against the accepted failure rate in the network, including the
probability of other failures that may occur in the network during its operational lifetime (e.g., failures
due to re-work or re-configuration in the link or due to other causes of cable or cabinet damage). In
assessing the result of this, the major question is whether single-mode fibres as specified in the
relevant Recommendations fulfil the requirements for a sufficiently long life time expectation. More
background is given to this question in this appendix.
Fibre storage at a certain radius in fibre management systems and in closures needs evaluation with
respect to fibre lifetime. For these applications, the loop size should be chosen to be large enough that
fibres are in the extrinsic region.
In many applications, loose fibre storage loops have a 30 mm radius with approximately 1-10 m of
fibre stored at a splice point. With improved macrobend fibre, as described in [ITU-T G.657], the size
of these loops could be reduced resulting in smaller enclosures but the amount of fibre required,
1-10 m, for splicing will likely remain the same. Smaller storage loops result in higher stress in the
fibre and thus potentially induce an increased risk of mechanical failure. Table I.1 below uses the
well-known power law theory of optical fibre reliability (see clause 8.2 and [b-IEC TR 62048]) to
show a 25-year failure probability as a function of loop size and fibre length, assuming a worst case
value for the dynamic stress corrosion susceptibility parameter nd = 18 as stated in [b-IEC 60793-2-
50]. Typical values of nd, which are greater than the specified minimum, produce lower calculated
failure probabilities than those in Table I.1. However, care should be taken that nd values are obtained
from the same test method [b-IEC 60793-1-33].
One can see from this example that using smaller coils will increase the failure probability. Typical
values of proof stress, which are greater than the specified minimum, produce lower calculated failure
probabilities than those in Table I.1. The differences in calculated failure probabilities with a variation
in proof stress levels are reduced as bend diameter is decreased.
In this particular network structure, a failure rate per individual single fibre cassette of 0.001% (105)
in 20 years will result in a 5% probability that in 20 years there will be one single spontaneous break
in the total network. This probability needs to be compared with the probability of other failures that
may occur in the distribution network during its 20-year operational lifetime. Such failures may be
due to re-work or re-configuration in the link or due to other causes of cable or cabinet damage. For
most access network situations, it may be assumed that the stated failure probability due to
spontaneous fibre breakage is much lower than the failure probability due to other causes. Each
operator has to determine the accepted failure rate based on more precise data on the outside plant
failure rate statistics.
Figure I.4 – Maximum storage length for a bent fibre and different values
of the dynamic stress corrosion parameter nd
I.5 Conclusion
The examples given support a 20-year operational lifetime for an appropriately installed network
equipped with bend-insensitive fibres as described in [ITU-T G.657], and bend radii less than 30 mm
with acceptable failure rates.
II.2 Factors having an impact on the objectives for the reliability of deployed optical
cables
II.2.2 The position of the optical cable in the telecommunication network
Tolerance for failure may not be equivalent throughout the optical network. A service provider must
consider several factors in making this decision. The table below gives an example of the different
role an optical cable can play in the various parts of an optical network. The table shows that
depending on the portion of the network being considered, a failure impacts different numbers of
users and may be more or less complicated to repair.
Series E Overall network operation, telephone service, service operation and human factors
Series F Non-telephone telecommunication services
Series Y Global information infrastructure, Internet protocol aspects and next-generation networks,
Internet of Things and smart cities
Printed in Switzerland
Geneva, 2016