ISO 18927:2002

INTERNATIONAL ISO
STANDARD 18927
First edition
2002-10-01
Imaging materials — Recordable compact
disc systems — Method for estimating the
life expectancy based on the effects of
temperature and relative humidity
Matériaux pour image — CD enregistrables — Méthode d'estimation de
l'espérance de vie basée sur les effets de la température et de l'humidité
relative

Reference number
ISO 18927:2002(E)
© ISO 2002

---------------------- Page: 1 ----------------------
ISO 18927:2002(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 copyright@iso.ch
Web www.iso.ch
Printed in Switzerland
©
ii ISO 2002 – All rights reserved

---------------------- Page: 2 ----------------------
ISO 18927:2002(E)
Contents Page
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Purpose and assumptions . 4
5 Measurements . 4
6 Accelerated stress test plan . 6
7 Data evaluation . 10
8 Disclaimer . 12
Annexes
A Numbering system for related International Standards. 13
B Ten-step analysis outline. 15
C Example of test plan and data analysis . 16
Bibliography. 27
©
ISO 2002 – All rights reserved iii

---------------------- Page: 3 ----------------------
ISO 18927:2002(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 3.
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.
Attention is drawn to the possibility that some of the elements of this International Standard may be the subject of
patent rights. ISO shall not be held responsible for identifying any or all such patent rights.
International Standard ISO 18927 was prepared by Technical Committee ISO/TC 42, Photography.
This International Standard is one of a series of standards dealing with the physical properties and stability of
imaging materials. To facilitate identification of these documents, they are assigned a number within the block from
18900-18999 (see annex A).
Annexes A, B and C of this International Standard are for information only.
©
iv ISO 2002 – All rights reserved

---------------------- Page: 4 ----------------------
INTERNATIONAL STANDARD ISO 18927:2002(E)
Imaging materials — Recordable compact disc systems — Method
for estimating the life expectancy based on the effects of
temperature and relative humidity
1 Scope
This International Standard specifies a test method for estimating the life expectancy of information stored on
recordable compact disc systems. Only the effects of temperature and relative humidity on the media are considered.
This International Standard does not cover the effects of light, air pollution, or time-dependent flow phenomena.
2 Normative references
The following normative documents contain provisions which, through reference in this text, constitute provisions of
this International Standard. For dated references, subsequent amendments to, or revisions of, any of these
publications do not apply. However, parties to agreements based on this International Standard are encouraged to
investigate the possibility of applying the most recent editions of the normative documents indicated below. For
undated references, the latest edition of the normative document referred to applies. Members of ISO and IEC
maintain registers of currently valid International Standards.
ISO/IEC 10149:1995, Information technology — Data interchange on read-only 120 mm optical data disks (CD-
ROM)
IEC 60908:1999, Audio recording — Compact disc digital audio system
Experimental statistics, US National Bureau of Standards Handbook 91, 1963
3 Terms and definitions
For the purposes of this International Standard, the following terms and definitions apply.
3.1
baseline
condition representing the disc at time of manufacture
NOTE This is customarily the initial parameter measurement taken prior to any application of stress. The designation is usually
for a stress time equal to zero hours.
t = 0
3.2
block error rate
BLER
ratio of erroneous blocks to total blocks measured at the input of the first (C1) decoder (before any error correction is
applied)
[IEC 60908:1999]
NOTE The more commonly reported value for BLER is the number of erroneous blocks per second measured at the input of the
C1-decoder during playback at the standard (1X) data rate.
©
ISO 2002 – All rights reserved 1

---------------------- Page: 5 ----------------------
ISO 18927:2002(E)
3.2.1
maximum block error rate
max BLER
maximum BLER measured anywhere on a disc
3.3
compact disc-recordable
CD-R
recordable optical disc in which information can be recorded to certain areas in compact disc format
NOTE 1 Information can be recorded one time and read many times.
NOTE 2 The term “compact disc-write once” (CD-WO) has also been used to describe this type of disc.
3.4
cumulative distribution function
F(t)
probability that a random unit drawn from the population fails by time, t, or the fraction of all units in the population
which fail by time, t
3.4.1
lognormal cumulative distribution function
F(t)
cumulative distribution function in which the logarithm of the relevant parameter, in this International Standard the
disc lifetime, has a normal distribution and is defined by the following equation:
� �
2
t
�
ln(x)−µ
l
1
−
1 1
2 σ
l
F (t)=√ e dx
σx
2π l
o
where
t is the time;
x
is a variable representing disc lifetime;
σ is the log standard deviation;
l
µ is the log mean;
l
ln(x) is the natural logarithm of x.
µ
l
NOTE When t = e , the lognormal cumulative distribution function evaluates to 0,5. In other words, the model predicts that half
the samples have failed at that time.
3.5
disc-at-once recording
method of recording a CD-R disc whereby the entire CD is recorded in one pass without turning off the laser
3.6
end-of-life
occurrence of any loss of information
3.7
extended-term storage conditions
storage conditions suitable for the preservation of recorded information having permanent value
©
2 ISO 2002 – All rights reserved

---------------------- Page: 6 ----------------------
ISO 18927:2002(E)
3.8
glass transition
reversible change in an amorphous polymer from, or to, a viscous or rubber condition to, or from, a hard and relatively
brittle one
3.8.1
glass transition temperature
T
g
approximate mid-point of the temperature range over which glass transition takes place
T
NOTE 1 can be determined readily only by observing the temperature at which a significant change takes place in a specific
g
electrical, mechanical, or other physical property.
NOTE 2 T can also be sensitive to the moisture content of the polymer.
g
3.9
information
signal or image recorded using the system
3.10
life expectancy
LE
◦
length of time that information is predicted to be acceptable in a system after dark storage at 21C5 and 0%RH
3.10.1
standardized life expectancy
SLE
minimum life span, predicted with 95 % confidence, of 95 % of the product stored at a temperature not exceeding
◦
25C5 and a relative humidity (RH) not exceeding 0%RH
3.11
retrievability
ability to access information as recorded
3.12
stress
experimental variable to which the specimen is exposed for the duration of the test interval
NOTE In this International Standard, the stress variables are confined to temperature and relative humidity.
3.13
survivor function
R(t)
probability that a random unit drawn from the population survives at least time, t, or the fraction of all units in the
population which survive at least time, t
NOTE R(t)= 1−F (t)
3.14
system
combination of material, hardware, software and documentation necessary for recording and/or retrieving
information
3.15
test cell
device that controls the stress to which the specimen is exposed
©
ISO 2002 – All rights reserved 3

---------------------- Page: 7 ----------------------
ISO 18927:2002(E)
3.16
track-at-once recording
method of recording a CD-R disc whereby each track is recorded individually with 150 empty sectors immediately
preceding it and two run-out sectors immediately following
3.17
uncorrectable error
error in the playback data that is not correctable by the cross interleave Reed-Solomon code defined in
IEC 60908:1999 as implemented in a system
4 Purpose and assumptions
4.1 Purpose
The purpose of this International Standard is to establish a methodology for estimating the life expectancy of
information stored on recordable compact disc systems. This methodology provides a technically and statistically
sound procedure for obtaining and evaluating accelerated test data.
The methodology deals only with the effects of temperature and humidity on the retrievability of stored information.
For this reason, this International Standard is primarily directed to those storage applications, e.g. libraries and
archives, in which exposure to other influences potentially detrimental to information life expectancy, such as
chemical agents, intense light sources and improper handling, is controlled and minimized.
4.2 Assumptions
The validity of the procedure defined by this International Standard relies on three assumptions:
— specimen life distribution is appropriately modelled by the lognormal distribution;
— the kinetics of the dominant failure mechanism is appropriately modelled by an Eyring acceleration model;
— the dominant failure mechanism acting at the usage condition is the same as that at the accelerated conditions.
Publications by Hamada and Stinson provide data indicating that these assumptions are applicable to compact disc-
recordable (CD-R) systems (see references [6] and [7] in the Bibliography).
5 Measurements
5.1 Summary
A sampling of eighty recorded discs shall be divided into five groups according to a specified plan. Each group of
discs (test cell) shall be subjected to one of five test stresses, combinations of temperature and relative humidity.
Periodically during the stress conditions, all discs from each stress group shall have their block error rate (BLER)
measured. Data collected at each time interval for each individual disc are then used to determine a lifetime for that
disc.
The disc lifetimes at each stress level are fitted to a lognormal distribution to determine a mean lifetime for the stress.
The resulting five mean lifetimes are regressed against temperature and relative humidity according to an Eyring
acceleration model. This model is then used to estimate the distribution of lifetimes at a usage condition.
5.2 BLER
End-of-life is the occurrence of any loss of information. Ideally, each specimen is tested until the first loss of
information occurs. Realistically, this is impractical. This International Standard considers max BLER to be a high-
level estimate of the performance of the system. The objective of measuring BLER is to establish a practical
©
4 ISO 2002 – All rights reserved

---------------------- Page: 8 ----------------------
ISO 18927:2002(E)
estimation of the system's ability to read recorded data without uncorrectable errors. A change in max BLER in
response to the time at an accelerated temperature and humidity is the principal quality parameter.
−2
IEC 60908:1999 states that the BLER averaged over any 10 s shall be less than 3× 10 . At the standard (1X) data
rate, the total number of blocks per second entering the C1-decoder is 7 350. Thus, an equivalent limit on BLER is
220 block per second.
For the purpose of this International Standard, recorded data are considered to have reached end-of-life when the
BLER, measured as erroneous blocks per second, exceeds 220 anywhere on the disc, i.e. when the max BLER
exceeds 220. It is recognized that the correlation between actual loss of information and max BLER is very system
dependent. A BLER of 220 is an arbitrary level chosen as a predictor of the onset of uncorrectable errors and thereby
end-of-life.
5.3 Test equipment
5.3.1 General requirements
A compact disc player that conforms to ISO/IEC 10149 and software capable of producing a display of max BLER.
If it becomes necessary to replace the test equipment, the USA National Bureau of Standards Handbook 91 shall be
followed for correlating test equipment outputs.
The make, model and version of the test equipment (including software) shall be reported with the test results.
5.3.2 Calibration and repeatability
A calibration, according to the tester manufacturer's procedure, shall be performed prior to any measurement data
being collected. A calibration disc shall be available from an accredited source.
In addition to the calibration disc, one control disc shall be maintained at ambient conditions and its max BLER
measured before and after each data collection interval. A control chart shall be maintained for this control disc.
The mean and standard deviation of the control disc shall be established by collecting at least five measurements.
Should any individual max BLER reading differ from the mean by more than three times the standard deviation, the
problem shall be corrected and all data collected since the last valid control point shall be remeasured.
5.4 Test specimen
5.4.1 General requirements
A test specimen is any disc that, after recording, meets ISO/IEC 10149 specifications and contains representative
data extending to within 2mm of the maximum recording diameter.
5.4.2 Specimen selection
All discs shall be nominally identical with regard to substrate groove structure, layer structure, coating composition,
recording capacity, and age prior to test initiation. It is preferred that the CD-R media be chosen from different lots
and production lines in order to be representative of normal process variations.
All discs shall be maintained in the manufacturer's transportation and storage conditions prior to recording.
The nominal disc capacity shall be reported with the test results.
©
ISO 2002 – All rights reserved 5

---------------------- Page: 9 ----------------------
ISO 18927:2002(E)
5.4.3 Recording system
Specimen discs may be recorded in any appropriate recording device. Since the extrapolated lifetime is a function of
the system including the CD-R media, all discs shall be recorded identically to the extent possible. Similar recording
devices shall be used as well as similar software and recording conditions. Discs recorded on different physical
devices shall be distributed as equally as possible across the test cells.
The make and model of the recording device, the linear velocity employed during recording and the software used in
the recording system shall be reported with the test results.
5.4.4 Ambient recording conditions
Ambient conditions during recording shall be within the following limits:
◦ ◦
—Temperature: 15C3to 5 C;
—Relative humidity: 45 %to .75 %
During recording, the recording system shall be isolated from external vibrations.
5.4.5 Recording method
It is strongly recommended that all discs be recorded using the “disc-at-once” method. If discs are recorded using the
“track-at-once” method, all errors occurring at the gap between tracks shall be ignored for the purpose of this
International Standard. Packet writing (in which several write events are allowed within a track) shall not be
employed.
Independent of the writing method, the specimen discs shall be recorded as a single session and finalized.
6 Accelerated stress test plan
6.1 General
Information properly recorded in a CD-R system of good manufacture should have a life of several years or even
decades. Consequently, it is necessary to conduct accelerated ageing studies in order to develop a life expectancy
estimate. The key is conducting a test plan that will provide the information necessary to satisfactorily evaluate the
particular system.
Many accelerated life test plans follow a rather traditional approach in specimen selection, experimentation and data
evaluation. These traditional plans share the following characteristics:
a) The total number of specimens is evenly divided amongst all the accelerated stresses.
b) Each stress is evaluated at the same time increments.
c) The Arrhenius relationship is used as the acceleration model.
d) The least squares method is used for all regressions.
e) The calculated life expectancy is for the mean or median life rather than for the first few failure percentiles.
On the other hand, optimum test plans have been proposed which differ in significant aspects from traditional plans.
These plans have the following characteristics:
— two and only two acceleration levels for each stress;
— a large number of specimens distributed mostly in the lowest stress levels;
— the need to know the failure distribution, a priori, in order to develop the plan.
©
6 ISO 2002 – All rights reserved

---------------------- Page: 10 ----------------------
ISO 18927:2002(E)
The maximum effectiveness of a plan can either be estimated before the test starts or determined after the results
have been obtained. As each CD-R system will have different characteristics, a specific, detailed optimum plan is
impossible to forecast.
This test plan borrows from the optimum plan, the traditional plan and previous experience with the systems, test
equipment and accelerated test stresses to put together a compromise test plan. Modifications of this plan will be
required to design the best plan for other applications. The methodology shall be applicable to all CD-R media
assessments.
6.2 Stress conditions
6.2.1 General
As mentioned in 6.1, an optimum test plan utilizes only two stress levels for each parameter evaluated. This is
because, in an ideal case, the relationship between changes in the parameter investigated and changes in stress are
known. The compromise test plan, documented in this International Standard, does not make such an assumption;
therefore, three different stress levels per parameter shall be used so that the linearity of the parameter function
versus the stress level may be demonstrated.
The test plan shall have the majority of test specimens placed at the lowest stress condition. This minimizes the
estimation error at this condition and results in the best estimate of the degradation rate at a level close to the usage
condition. The greater number of specimens at the lower stress condition also tends to equalize the number of
failures observed by test completion.
For implementing the test plan documented in this International Standard, five stress conditions shall be used. The
minimum distribution of specimens among the stress conditions that shall be used is shown in Table 1. Additional
specimens and conditions may be used if desired for improved precision.
Table 1 — Summary of stress conditions
Minimum
Test cell Number of Incubation Minimum total Intermediate
Test stress equilibration
number specimens duration time RH
duration
T /RH hh RH h
inc inc inc
◦
1 80 C/85 % RH 10 500 2 000 31 % 6
◦
2 80 C/70 % RH 10 500 2 000 31 % 5
◦
3 80 C/55 % RH 15 500 2 000 31 % 4
◦
70 C/85 % RH 33 %
4 15 750 3 000 8
◦
60 C/85 % RH 36 %
5 30 1 000 4 000 11
6.2.2 Temperature ()T
The temperature levels chosen for this test plan are based on the following:
— there shall be no change of phase within the test system over the test-temperature range. This restricts the
◦ ◦
temperature to greater than 0 C and less than 100 C;
— the temperature shall not be so high that plastic deformation occurs anywhere within the disc structure.
◦
The typical substrate material for CD-R media is polycarbonate (glass transition temperature ∼ 150 C). The glass
transition temperature of other layers may be lower. Experience with high-temperature testing of CD-R discs
◦
indicates that an upper limit of 80 C is practical for most applications.
6.2.3 Relative humidity (RH)
Practical experience shows that 85 % RH is the upper limit for control within most accelerated test cells. This is due
to the tendency for condensation to occur on cool sections of the chamber, e.g. observation windows, cable ports,
©
ISO 2002 – All rights reserved 7

---------------------- Page: 11 ----------------------
ISO 18927:2002(E)
wiper handles, etc. The droplets may become dislodged and entrained in the circulating air within the chamber. If
these droplets fall on the test specimen, false error signals could be produced.
6.2.4 Rate of stress change
The process described in this International Standard requires that discs undergo a transition from the ambient
conditions to stress conditions and back again a number of times during the course of testing. The transition (or
ramp) duration and conditions shall be chosen to allow sufficient equilibration of absorbed substrate moisture.
Large departures from equilibrium conditions may result in the formation of liquid water droplets inside the substrate
or at its interface with the information-recording layer. Gradients in the water concentration through the thickness of
the substrate shall also be limited. These gradients drive expansion gradients which can cause significant disc
curvature.
In order to minimize the effects of moisture-concentration gradients, the ramp profile outlined in Table 2 shall be
used. The objects of the profile are:
— to avoid any situation that may cause moisture condensation within the substrate;
— to minimize the time during which substantial moisture gradients exist in the substrate;
— to produce at the end of the profile a disc that is sufficiently equilibrated to proceed directly to testing without
delay.
Table 2 — Temperature and relative humidity transition (ramp) profile
Process step Temperature Relative humidity Duration
◦
C %h
Start at atT RH —
amb amb
TT, RH ramp to to RH 1,5± 0,5
inc int
RH ramp at T to RH 1,5± 0,5
inc inc
Incubation at aT t RH See Table1
inc inc
RH ramp at T to RH 1,5± 0,5
inc int
Equilibration at atT RH See Table1
inc int
TT, RH ramp to to RH 1,5± 0,5
amb amb
End at atT RH —
amb amb
NOTE 1 T and RH are room ambient temperature and relative humidity; T and RH are the stress incubation temperature and
amb amb inc inc
relative humidity; and RH is the intermediate relative humidity that at T supports the same equilibrium moisture absorption in
int inc
polycarbonate as that supported at T and RH (see Table 1).
amb amb
◦ ◦
NOTE 2 Transitions should not deviate from a linear change over the chosen duration by more than ± 2 C and ± 3 CRH. Ramp
transitions may be controlled automatically or manually.
The profile accomplishes this by varying the moisture content of the disc only at the stress incubation temperature,
and allowing sufficient time for equilibration during ramp-down based on the diffusion coefficient of water in
polycarbonate.
Figure 1 graphically portrays the temperature and humidity changes that would occur during one cycle of incubation
◦
80C85%RH
at and .
6.2.5 Independent verification of chamber conditions
A system independent of the chamber control system shall be used to monitor temperature and relative humidity
conditions in the test chamber during the stress test.
©
8 ISO 2002 – All rights reserved

---------------------- Page: 12 ----------------------
ISO 18927:2002(E)
◦
Figure 1 — Graph of nominal 80 C/85 % RH transition (ramp) profile (as per Tables 1 and 2)
6.2.6 Specimen placement
Disc specimens shall be placed uncovered, either vertically or horizontally, within the test chamber. Discs shall be
aligned so that their surface is parallel to the chamber airflow. A space of at least 2mm shall be maintained between
discs.
6.2.7 Other influences
During the course of the stress test, the discs shall be shielded from excessive illumination and potentially corrosive
fumes, gases and liquids.
6.3 Accelerated test cell sample population
In order to estimate the log mean and log standard deviation of a lognormal distribution, at least 10 failures shall be
◦
observed. Observing at least 10 failures is generally not a problem for a realistic test time at 80 C/85 % RH, but
becomes more difficult at milder stress temperature and relative humidity combinations. Assigning a larger
percentage of the specimens to the lower stresses increases the chance of observing the necessary number of
failures within a practical time interval.
Specimens that have not failed at the end of the test duration shall be time censored. This is also known as Type I
censoring (see reference [8], page 233, in the Bibliography).
To compute the estimated failure time for each disc, it is necessary to first determine a transformation of max BLER,
such as ln (max BLER), that results in a linear time dependence. Standard linear regression techniques shall be
used to find the best fit to the transformed data. The failure time for each disc shall then be computed by interpolation
using each disc's regression equation.
If 10 failures are not observed by the end of the test duration, then failures may be estimated by extrapolation.
6.4 Time intervals
For data collection, measurement of max BLER shall occur at four time intervals for each disc. The baseline
measurement (at test initiation) shall be an additional data point. Within a stress condition, the intervals shall be
©
ISO 2002 – All rights reserved 9

---------------------- Page: 13 ----------------------
ISO 18927:2002(E)
constant. At the milder stress conditions, the intervals are longer. The longer time intervals provide the opportunity for
more failures to occur at the milder stress conditions.
6.5 Test plan
Table 1 specifies the temperatures, relative humidities, time intervals, minimum total test time and specimen
distributions for each stress condition. A separate group of specimens is used for each stress condition. This
constitutes a constant stress test plan.
◦
All temperatures have an allowed range of ± 2 C; all relative humidities have an allowed range of ±3%RH.
◦
The intermediate relative humidity (RH ) in Table 1 is calculated assuming 25C5 and 0%RH ambient conditions.
int
If the ambient is different, the intermediate relative humidity to be used is calculated using the equation:
0,24+ 0,003 7×T
amb
RH = × RH
int amb
0,24+ 0,003 7×T
inc
◦
where T and T are respectively the ambient and incubation temperature in units of CR, and H is the
amb amb
inc
ambient relative humidity.
The stress conditions tabulated in Table 1 offer sufficient combinations of temperature and relative humidity to satisfy
the mathematical requirements of the Eyring model (see 7.2) to demonstrate linearity of either max BLER or
ln (max BLER) versus time, and t
...