ISO/TR 18945:2018

Imaging materials - Pictorial colour reflection prints - Comparison of image degradation observed between ISO 18930 accelerated weathering test method and outdoor exposure

This document describes the experimental framework, results, and conclusions from a round robin test that was performed in order to establish correlations between accelerated weathering according to the ISO 18930 test method and outdoor weathering at nine outdoor sites. The types of digital printing technology that were used in this round robin test are aqueous inkjet, solvent inkjet, UV curable inkjet, digitally-exposed silver halide, and thermal mass transfer. The image print stability data and correlations of this document are to be considered illustrative of the performance of these classes of materials. Extension of these correlations to other classes of materials, such as dye sublimation, is verified by appropriate experimentation.

Matériaux pour l’image — Réflexion des impressions photographiques en couleurs — Comparaison de la dégradation de l’image observée entre la méthode d’essai de vieillissement accéléré de l’ISO 18930 et l’exposition extérieure

General Information

• Status: Published
• Publication Date: 22-Oct-2018
• ICS: 37.040.99 - Other standards related to photography
• Technical Committee: ISO/TC 42 - Photography
• Drafting Committee: ISO/TC 42/WG 5 - Physical properties and image permanence of photographic materials
• Current Stage: 6060 - International Standard published
• Start Date: 23-Oct-2018
• Completion Date: 13-Dec-2025

Overview

ISO/TR 18945:2018 is a Technical Report from ISO that documents a round‑robin study comparing image degradation observed in ISO 18930 accelerated weathering tests with actual outdoor exposure at nine sites. It covers the experimental framework, data analysis, and conclusions about how well accelerated laboratory simulations predict real‑world fading and colour shifts of pictorial colour reflection prints. The tested digital printing classes include aqueous inkjet, solvent inkjet, UV‑curable inkjet, digitally‑exposed silver halide, and thermal mass transfer. Correlations presented are illustrative for these classes and note that extension to other technologies (e.g., dye sublimation) requires verification.

Key Topics

• Test scope and methodology
  • Round‑robin design comparing ISO 18930 accelerated tests with outdoor racks at nine geographic sites.
  • Both outdoor weathering and laboratory accelerated weathering procedures are described.
• Acceleration and correlation metrics
  • Acceleration factor (time ratio between outdoor and accelerated tests).
  • Pearson correlation coefficient to quantify linear agreement between test methods.
• Fundamental drivers of degradation
  • Importance of light, heat, and water; ultraviolet (295–400 nm) is especially damaging to polymers and colourants.
  • Role of spectral power distribution (SPD) of test lamps; filtered xenon‑arc sources are highlighted as state‑of‑the‑art for simulating sunlight.
• Data analysis and variability
  • Colour fade acceleration factors, colour shift analysis, replicability of results, and two‑year data evaluation.
  • Cautions on site‑ and material‑specific variability (acceleration factors can vary substantially; one study noted ±50% variation by material at a location).
• Supporting material
  • Informative annexes with SPD data, photographs of degradation, environmental condition datasets, and detailed comparisons (e.g., Annex G).

Applications

ISO/TR 18945:2018 is practical for:

• Material and ink manufacturers assessing print stability and comparing technologies.
• Testing laboratories developing accelerated weathering protocols and interpreting correlations to outdoor life.
• Print service providers, signage, and outdoor graphics producers estimating outdoor durability and warranty lifetimes.
• Conservators and photographic archivists wanting evidence‑based guidance on expected outdoor behaviour.
• R&D teams selecting substrates/coatings or validating new printing processes against outdoor performance.

Practical takeaways: use ISO 18930 accelerated tests to shorten evaluation time, but apply acceleration factors cautiously and validate correlations for each material class and exposure climate.

Related Standards

• ISO 18930 (accelerated and outdoor weathering test methods for print stability)
• ISO 4892‑1 (general guidance on laboratory weathering devices)
• ASTM G155 (xenon arc accelerated weathering) These references help implement and interpret accelerated weathering in the context of ISO/TR 18945 findings.

Standards Content (Sample)

TECHNICAL ISO/TR
REPORT 18945
First edition
2018-10
Imaging materials — Pictorial colour
reflection prints — Comparison of
image degradation observed between
ISO 18930 accelerated weathering test
method and outdoor exposure
Matériaux pour l’image — Réflexion des impressions photographiques
en couleurs — Comparaison de la dégradation de l’image observée
entre la méthode d’essai de vieillissement accéléré de l’ISO 18930 et
l’exposition extérieure
Reference number
©
ISO 2018
© ISO 2018
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting
on the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address
below or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: [phone]
Fax: [phone]
Email: [e-mail]
Website: www.iso.org
Published in Switzerland
ii © ISO 2018 – All rights reserved

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 General considerations for accelerated weathering tests . 2
5 Materials . 4
6 Test methods . 5
6.1 Outdoor exposure tests . 5
6.2 Laboratory accelerated weathering tests . 5
6.3 Data analysis and work-up . 5
7 Results and discussion . 6
7.1 Colour Fade Acceleration Factors . 6
7.2 Replicability of data . 7
7.3 Applicability to multiple digital printing technologies . 8
7.4 Effects of colour and patch darkness . 9
7.5 Analysis of colour shifts . 9
7.6 Two-year data analysis .10
7.7 Correlation coefficients and predictive correlations .11
7.8 Example — Degradation of Material H4 .12
7.9 Comparison of material degradation during outdoor and ISO 18930 accelerated
laboratory weathering tests (see Annex G) .14
7.9.1 General.14
7.9.2 Colour fade graphs .14
7.9.3 Comparison of ISO 18930 accelerated tests to nine outdoor exposure sites .15
7.9.4 Colour shift graphs .15
8 Conclusions and recommendations .15
Annex A (informative) Spectral power distribution for accelerated laboratory weathering tests .16
Annex B (informative) Photographs of weathered test target degradation .17
Annex C (informative) Comparison of accelerated weathering test methods and outdoor results .21
Annex D (informative) The various types of deterioration observed in ISO 18930 .28
Annex E (informative) Effects of the angle of inclination in outdoor testing.30
Annex F (informative) Environmental condition data under real outdoor conditions .38
Annex G (informative) Comparison of material degradation during outdoor and ISO 18930
accelerated laboratory weathering tests .42
Bibliography .93
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.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular the different approval criteria needed for the
different types of ISO documents should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2 (see www .iso .org/directives).
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. Details of
any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www .iso .org/patents).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation on the voluntary nature of standards, the meaning of ISO specific terms and
expressions related to conformity assessment, as well as information about ISO's adherence to the
World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT) see the following
URL: www .iso .org/iso/foreword .html.
This document was prepared by Technical Committee ISO/TC 42, Photography.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www .iso .org/members .html.
iv © ISO 2018 – All rights reserved

Introduction
Printed digital images are used in many applications in which they are exposed to outdoor weathering.
ISO 18930 provides standardized test procedures to evaluate image stability both in real-time outdoor
weathering tests and in accelerated laboratory simulations of the weathering process. Accelerated
laboratory weathering tests have been developed as a result of the desire to obtain test results faster
than would be obtained by actual outdoor exposure. However, accelerated weathering tests only have
value if they can be correlated with actual outdoor performance.
TECHNICAL REPORT ISO/TR 18945:2018(E)
Imaging materials — Pictorial colour reflection prints —
Comparison of image degradation observed between ISO
18930 accelerated weathering test method and outdoor
exposure
1 Scope
This document describes the experimental framework, results, and conclusions from a round robin test
that was performed in order to establish correlations between accelerated weathering according to the
ISO 18930 test method and outdoor weathering at nine outdoor sites.
The types of digital printing technology that were used in this round robin test are aqueous inkjet,
solvent inkjet, UV curable inkjet, digitally-exposed silver halide, and thermal mass transfer. The
image print stability data and correlations of this document are to be considered illustrative of the
performance of these classes of materials. Extension of these correlations to other classes of materials,
such as dye sublimation, is verified by appropriate experimentation.
2 Normative references
There are no normative references in this document.
3 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
ISO and IEC maintain terminological databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https: //www .iso .org/obp
— IEC Electropedia: available at http: //www .electropedia .org/
3.1
digital printing media
recording elements used by digital printers to receive inks or pre-formed colourants
EXAMPLE The substrate may be paper, plastic, canvas, fabric, metal, or other ink-receptive material; the
substrate may, or may not, be coated with an ink-receptive layer. The category of digital printers includes inkjet,
electrophotographic, and thermal transfer.
3.2
laminate
overlaminate
layer of material that goes over the top or bottom of a specimen
Note 1 to entry: Usually to provide water-resistance, physical, and/or ultraviolet (UV) light protection of the
specimen during a weathering test. A layer of protective film is applied with a pressure-sensitive or heat-
activated adhesive.
3.3
accelerated laboratory weathering
simulated weathering where instruments (weathering devices) are used to obtain very controlled
conditions that simulate, to some degree, and generally accelerate, the outdoor weathering results
[2] [16]
Note 1 to entry: The use of such instruments is described in ISO 4892-1 and ASTM G151 .
3.4
outdoor weathering
actual placement of specimens outdoors in specific locations
Note 1 to entry: This is differentiated from simulated weathering where instruments (weathering devices) are
used to obtain very controlled conditions that simulate, to some degree, and generally accelerate the outdoor
[2] [16]
weathering results. Use of such instruments is described in ISO 4892-1 and ASTM G151 .
3.5
reciprocity failure
non-equivalence in weathering results between a long exposure/low-intensity experiment and its short
exposure/high-intensity counterpart with an equivalent intensity-time product
3.6
daylight filter
optical filter or combination of filters that modifies the spectral power distribution of a light source to
better represent some defined daylight spectrum
Note 1 to entry: These filters are not related to the blue filters used in the photographic industry for the change
of correlated colour temperature of light sources.
[5]
Note 2 to entry: Adapted from ISO 18913 .
3.7
coefficient of variation
standard deviation of a variable divided by the arithmetic mean of the variable
3.8
Pearson correlation coefficient
statistical measure of the degree of linear correlation between two variables, with value between −1,0
and +1,0 inclusive, where a value of +1,0 represents perfect positive correlation, a value of 0,0 signifies
no correlation, and a value of −1,0 represents perfect negative correlation
3.9
acceleration factor
ratio of the time required to reach an endpoint in an outdoor weathering test to the time required to
reach the same endpoint in a laboratory accelerated weathering test
3.10
colour fade acceleration factor
acceleration factor for which the bases of comparison are the ratios of reflected optical density during
the test to the initial reflected optical density prior to the start of the test
4 General considerations for accelerated weathering tests
The ability to accurately predict the long-term outdoor performance of materials and printed images is
essential to many industries. Since many of the relevant products are designed to last years or decades,
accelerated weathering test methods have been developed to more rapidly assess outdoor performance
and to investigate failure mechanisms associated with outdoor exposure. Unfortunately, this is an
extremely complex task.
The three key components of accelerated weathering tests are heat, light, and water. The primary
determinant of the degree of correlation for between outdoor weathering and an accelerated test
method is the degree to which the spectral power distribution (SPD) of the light source in the test
[7]
chamber matches the SPD of sunlight . This is so critical because material photodegradation
[8]
mechanisms are very specific to certain wavelengths of light . The UV spectrum between 295 nm and
400 nm is responsible for most of the damage to polymers and colourants. The current state-of-the-
art light source is filtered xenon arc lamps. In a comprehensive study of the accelerated weathering of
[9]
polyester gel coats, Crump found that xenon arc weathering gave higher correlation coefficients than
2 © ISO 2018 – All rights reserved

[10][11]
methods employing carbon arc or fluorescent light sources. Previous investigations by Klemann
also indicated high correlation coefficients for xenon arc light sources.
Water exposure is also essential because many materials exhibit hydrolytic degradation pathways.
Heat, in terms of elevated chamber temperatures, is used to accelerate all of the reactions that
contribute to material and image degradation. Other factors such as ozone, pollutants, freeze-thaw
cycles, and abrasion due to airborne particles may also affect material longevity, but are not included in
most accelerated test cycles.
Two metrics are used to gauge the efficacy of accelerated weathering test methods: the acceleration
factor and the Pearson correlation coefficient. An acceleration factor is a scale factor that relates the
rate of degradation in an accelerated test to the rate of degradation in real-time outdoor exposure. For
example, if a colour patch fades by 40 % over one year on an outdoor rack in South Florida and also fades
40 % after 1 month of an accelerated weathering test, then the acceleration factor would be 12, as one
month of accelerated testing is equivalent to 12 months of outdoor testing. The correlation coefficient
is the degree to which, and the consistency of, the agreement between accelerated and outdoor testing.
The user of any accelerated weathering method should be cautioned that the acceleration factors are
specific to both the outdoor location and to the material, or combination of materials, that are tested. It
should be obvious that acceleration factors depend upon the climate of the outdoor site. Average radiant
exposure, rainfall, relative humidity, and temperatures of an outdoor location all affect the acceleration
factor. Indeed, even year to year climatic variations will change the acceleration factor to some degree.
What may be less obvious is that there are also some differences in acceleration factor for different
materials. This is due to the different photodegradation mechanisms and their wavelength specificity,
to the rates of water absorption and the saturation moisture levels, and to any changes in degradation
mechanisms as a function of temperature (for example, outdoor conditions are below a polymer
glass transition temperature and the temperature of an accelerated weathering test is above it). An
investigation of fade of colour patches on signs and labels showed that that the average acceleration
[10]
factor for a set location may vary as much as ±50 % by material construction .
NOTE If use of an acceleration factor is desired in spite of the warnings given in this document, such
acceleration factors for a particular material are only valid if they are based on data from a sufficient number
of separate exterior or indoor environmental tests and accelerated laboratory exposures so that results used to
relate times to failure in each exposure can be analysed using statistical methods, see ISO 4892-1.
No standard accelerated weathering test method results in a perfect correlation with outdoor
[17]
performance. ASTM G155 , Cycle 1, and its predecessor ASTM G26, uses one or more xenon lamps
with borosilicate type S inner and outer filters, which gives an excellent approximation for the SPD
[12]
of sunlight, and has a periodic water spray, but is an isothermal test. SAE J2527 test cycle and its
predecessor SAE J1960 both include segments with high temperatures and a segment with lower
temperature, water spray, and no light, to simulate night. For some materials that are sensitive to
expansion and contraction, or to the stresses of drying while heating, this type of day-night cycle may
give more realistic results. However, the quartz inner/borosilicate outer filter combination of these SAE
tests exposes samples to light in the 280 nm to 295 nm range that would be screened out by the earth’s
ozone layer outdoors.
To improve upon previous accelerated weathering standards, the ISO 18930 test method was
developed. It was confirmed in 2011. The light source SPD (see Annex A) is specified in terms of
spectral output by 10 nm or 20 nm bands of wavelengths so as to provide a best match to the SPD of
[18]
sunlight in CIE 85:1989 , Table 4. Four cycle segments are incorporated: three at high black panel and
chamber temperatures with light exposure, and one at lower temperature in the dark. Water spray is
included for one of the high-temperature segments and for the cool, dark cycle segment (Table 1). The
International Standard requires that testing be conducted at a 45° angle of inclination, although other
[13]
angles of inclination may be added, as this maximizes the solar irradiance received by the samples .
Table 1 — ISO 18930 xenon arc exposure test cycle
Cycle Time Irradiance – Irradiance – Black Panel Chamber Relative Water
Segment (min) Narrowband Broadband (300 Temperature Temperature Humidity Spray
(340 nm) to 400 nm)
2 2
W/m W/m °C °C %
1 40 0,55 ± 0,02 60 ± 2 63 ± 2 40 ± 2 50 ± 6 None
2 20 0,55 ± 0,02 60 ± 2 — 40 ± 2 — Front
3 60 0,55 ± 0,02 60 ± 2 63 ± 2 40 ± 2 50 ± 6 None
4 60 0,00 0 — 38 ± 2 — Front
This paper describes the details and results of a round-robin study with nine outdoor global locations
and six laboratories running ISO 18930 in order to validate the new test method.
5 Materials
This investigation encompassed 32 material/ink combinations and digital printing technologies.
Technologies represented included aqueous inkjet, solvent inkjet, UV inkjet, digital silver halide,
thermal transfer, and for comparison, flexography. Some were overlaminated, others remained directly
exposed to the elements. For all materials, two replicates of the target below were printed. The target
has six patches each varying in lightness for cyan, magenta, yellow, true black, red, green, blue, and
process black (CMY) see Figure 1. Two small white patches were included for measurements of material
yellowing, and large black and white patches were added below for gloss measurements.
Figure 1 — ISO 18930 Round Robin Test Target
Two replicates each of the test targets were printed for nine outdoor sites and six accelerated test
instruments running ISO 18930 (see Table 1). After printing, the samples were maintained at 23 °C and
50 % relative humidity until the start of the tests.
4 © ISO 2018 – All rights reserved

6 Test methods
6.1 Outdoor exposure tests
Both accredited and non-accredited outdoor sites were included in this investigation (see Table 2).
Printed test targets were mounted on aluminum panels. These outdoor panels were placed on racks at
a 45° angle of inclination, south facing. For a comparison of exposures at angles of inclination of 45° and
90°, see Annex E. Measurements of the colour patches were taken at 0 year, 1 years and 2 years for the
outdoor sites.
Table 2 — Outdoor test site climate data
Site Latitude Radiant Precipitation Average Accredited
Exposure Temperature Lab
MJ/m2/y mm °C
South Florida, USA 25,87° N 6 588 1 655 23 YES
San Diego, CA USA 33,03° N 6 602 262 18 NO
DSET, Arizona, USA 33,90° N 8 004 255 22 YES
Tokyo, Japan 35,71° N 4 959 1 682 14 YES
Chicago, IL USA 41,78° N 5 100 856 10 YES
Sanary, France 43,13° N 5 500 700 13 YES
Milwaukee, WI USA 43,14° N 5 103 884 9 NO
Marly, Switzerland 46,78° N 4 590 1 075 9 NO
Mortsel, Belgium 51,17° N 3 708 825 10 NO
6.2 Laboratory accelerated weathering tests
All accelerated weathering instruments were set for Borosilicate Type S inner/Borosilicate Type S
outer, Daylight Q, Daylight B/B, Quartz/#295, or other combinations appropriate to match the SPD
requirements associated with ISO 18930. Colour measurements were taken after 0 h, 24 h, 200 h,
400 h, 800 h, 1 200 h, 1 600 h and 2 000 h of exposure to the ISO 18930 test cycle (see Table 1) for all
accelerated testing chambers. For some of the chambers, the testing time was increased to as long as
5 200 h of exposure. For all colour measurements 45°/0° geometry, a 10° observer, and D65 illuminant
were specified. Spectrophotometer data was converted to reflected optical densities according to the
ANSI Status A Standard for densitometer filters.
6.3 Data analysis and work-up
The procedure for data analysis employed optical density ratios – the ratio of optical density to initial
optical density. For primary colours only a single density ratio was tracked. For secondary colours
two density ratios were tracked, and for the process black patches all four densities (C, M, Y, and K
or D ) were tracked. For the secondary and process patches, the difference in density ratios for the
VIS
relevant densities were also tracked as a measure of colour shifts. For the two white patches, ΔE76 was
measured to evaluate substrate yellowing.
For each outdoor site after one year of exposure the test targets were measured with the
spectrophotometer. Density ratios were calculated and used on a patch by patch basis. To find the
acceleration factor for a single patch in an accelerated weathering chamber, the number of hours
needed to obtain the same density as that of the outdoor site were determined via linear interpolation
of the accelerated colour data. The acceleration factor is then calculated as 8 766 h (one year) divided by
the number of hours in ISO 18930 that gives the same density ratio. The 48 patch acceleration factors
on a target could then be used for statistical comparisons by material, accelerated testing laboratory,
outdoor site, print technology, etc. Only density ratios between 0,95 and 0,30 were used for analysis, as
it was thought that less than a 5 % density loss did not give a large enough signal to noise ratio, and that
degradation would slow down or even reach an asymptote at density ratios less than 0,30.
7 Results and discussion
7.1 Colour Fade Acceleration Factors
Acceleration factors were calculated for colour fade, colour shifts, and for background yellowing. For
reasons that will be specified later, colour fade acceleration factors were found to be the most useful
output of the study.
Initially, the accelerated ISO 18930 tests were scheduled to run only 2 000 h. However, it was soon
determined that this test duration was insufficient, especially when correlating to the more aggressive
climates of South Florida, Arizona, and San Diego. This was found to be critical in the determination of
correct acceleration factors. Not all 48 patches on a test target yield useful data points, and these data
points are first available when the patch on the accelerated test target reaches the same density ratio as
the outdoor test patch. There are two possibilities for missing data points:
a) The outdoor test patch has a density ratio above 0,95 or below 0,30 and is excluded from analysis;
b) The outdoor test patch is in the correct density ratio range, but the accelerated test has not been
run long enough to reach that density ratios.
In Case 1, the data points will never be available. For Case 2, however, more data points come in as the
length of the accelerated test is extended. This causes the apparent acceleration factor to decrease over
time until all of the Case 2 points come in and the apparent acceleration factor converges to the true
acceleration factor. An example of this is shown in Table 3 for South Florida, one of the most aggressive
climates.
Table 3 — Change in apparent acceleration factor as more accelerated test data is collected
Colour Patch Fade Data for South Florida Test Site
Hours of Accelerated Percentage of Maximum Apparent Acceleration Hours of Accelerated
Testing ISO 18930 Data Points Available Factor for 1 Year Testing ISO 18930
Outdoors
2 000 9 7,77 2 000
4 200 51 4,34 4 200
5 200 56 4,13 5 200
After accelerated testing was extended to 5 200 h to ensure that all of the obtainable data points
were collected, true acceleration factors could be determined for all nine outdoor sites. The average
acceleration factors for the 32 materials are shown in Table 4. As would be expected, the most aggressive
climates show smaller acceleration factors than the sites farther north; the trends intuitively seem to
make sense. The differences between the highest and lowest acceleration factors also scale with results
of previous studies that indicated approximately a factor of two ratio between South Florida and sites
[10][14]
with latitude of 42 N to 55°N .
Table 4 — Colour fade acceleration factors (AF) by site for 1 y outdoor exposure
Arizona Chicago Sanary, South San Tokyo, Mortsel, Marly,
FRA Florida Milwaukee Diego JP BEL CH
AF Colour
4,84 7,38 6,13 4,13 7,47 5,61 8,28 8,22 8,49
Fade – 1 year
Material Stdev
1,91 3,17 2,19 2,53 2,96 2,00 3,38 4,06 3,85
– 1 year
% Data Points
50 54 62 56 59 57 49 63 58
Available
6 © ISO 2018 – All rights reserved

7.2 Replicability of data
Consistency of the data is evaluated on a lab-lab basis in Table 5, for which the standard deviations
compare the acceleration factors for data points at a given lab to the average acceleration factor for that
material and outdoor location for labs 3 to 5. Note that only Labs 3, 4, and 5 were included, because the
other labs had different accelerated test durations. Replicability is evaluated in Table 6, for which the
standard deviations for the two replicates per lab are compared. In both Tables the standard deviations
were normalized to the coefficient of variation (standard deviation/average) for comparison. The
average coefficients of variation were 13,5 % and 14,6 % for lab-lab and replicate comparisons,
respectively. Since the lab-lab variation is barely higher that the variation of replicates in the same
instrument, in may be inferred that the standard test method is barely affected by changes of the test
instrument, as long as it is capable of meeting the specifications stipulated in the test method standard.
It is also seen that the variations are a bit lower for the more aggressive climates than for the higher
latitudes. This will be discussed later with the correlation coefficients.
Table 5 — Lab-lab coefficient of variation for colour fade failure hours
Outdoor Site Lab 3 Lab 4 Lab 5 Average
DSET Arizona 0,084 0,108 0,112 0,106
Chicago 0,131 0,125 0,172 0,146
Sanary, France 0,119 0,121 0,142 0,130
South Florida 0,064 0,087 0,063
Milwaukee, Wis. 0,171 0,219 0,170 0,191
San Diego 0,124 0,133 0,175 0,149
Tokyo, Japan 0,175 0,144 0,194 0,170
Mortsel, Belgium 0,175 0,156 0,215 0,184
Marly, Switzerland 0,114 0,157 0,165 0,153
Overall Average 0,137 0,136 0,159 0,146
Table 6 — Replicate coefficient of variation for colour fade failure hours
Outdoor Site Lab 1 Lab 2 Lab 3 Lab 4 Lab 5 Lab 6 All Labs
DSET Arizona 0,117 0,174 0,061 0,123 0,105 0,136 0,139
Chicago 0,130 0,141 0,076 0,114 0,128 0,180 0,137
Sanary, France 0,114 0,106 0,073 0,102 0,109 0,152 0,114
South Florida 0,103 0,164 0,061 0,104 0,102 0,152 0,127
Milwaukee, Wis. 0,139 0,137 0,071 0,292 0,141 0,179 0,163
San Diego 0,156 0,142 0,066 0,093 0,135 0,169 0,136
Tokyo, Japan 0,125 0,127 0,085 0,102 0,135 0,168 0,128
Mortsel, Belgium 0,154 0,122 0,076 0,116 0,086 0,156 0,132
Marly, Switzerland 0,164 0,160 0,078 0,148 0,141 0,189 0,156
Overall Avg. 0,130 0,140 0,074 0,126 0,125 0,164 0,135
7.3 Applicability to multiple digital printing technologies
The scope of ISO 18930 covers all digital printing. Since standards are generally developed to have
universal applicability across large classes of materials and technologies, materials and inks from five
digital printing technologies plus analog flexography were included in the round robin test. Table 7
shows a comparison of the acceleration factors for the different groups and confidence intervals of
±2σ around the averages. It is observed that the confidence intervals for the print/ink technologies are
larger than the differences between the averages of the technologies. So, with this small data set, it is
not possible to say that any of the print technologies show statistically significant differences. Indeed,
the only one for which that looks possible is digitally-exposed silver halide.
Table 7 — Colour fade acceleration factors by printing technology
Print Technology Average AF Stdev AF Lower Upper Materials
Confidence Confidence
Interval Interval
Aqueous Inkjet 5,32 2,22 0,88 9,76 11
UV Inkjet 4,91 1,46 1,99 7,83 8
Solvent Inkjet 5,83 2,24 1,35 10,31 6
Digital Silver Halide 8,80 1,94 4,92 12,68 3
Thermal Mass 5,21 0,99 3,23 7,19 2
Transfer
Flexography 5,96 — — — 2
Table 8 — Classification of test materials by print technology
Print Technology Number of Materials Materials Materials Inkjet Materials
Materials without with Pressure- with Liquid
Piezoelectric Thermal
Tested Protective Sensitive Clear Coats
Printheads Printheads
Layers Overlaminates
Aqueous Inkjet 11 8 3 0 7 4
UV Inkjet 8 7 0 1 8 0
Solvent Inkjet 6 3 3 3 6 0
Digital Silver Halide 3 0 3 0 — —
Thermal Mass 2 1 1 0 — —
Transfer
Flexography 2 1 1 0 — —
8 © ISO 2018 – All rights reserved

7.4 Effects of colour and patch darkness
The use of six patches for each colour of ascending colour density on the test target makes it simple to
break down the results by colour and patch darkness. The surprise here is that the average correlation
coefficient did vary to some extent with those two factors. While the rest of the colours are within 10 %
of the overall average, yellow patches faded 22 % faster than the average for all colours. There is also
a slight increase in acceleration factor as the patches in a series get lighter (from 6 to 1 in the target).
However, no satisfactory explanation for this behaviour has been proposed. Indeed, it is not known
whether these results are due to a few materials in a small data set, or whether they signify a real
phenomenon. See G.5 for an example of several materials for which the Tokyo data showed very high
acceleration factors, but the South Florida data was very close to the overall average acceleration factor.
Table 9 — Average colour fade AF by colour
Colour Average AF AF as % of
Overall Average
Cyan 5,25 80,8
Magenta 6,19 95,4
Yellow 7,97 122,7
Black (K) 5,96 91,7
Red 6,95 107,0
Green 6,67 102,8
Blue 6,11 94,2
Process Black 6,57 101,2
(CMY)
Overall Average 6,49 100,0
Table 10 — Average colour fade AF by patch darkness
Patch Average AF AF as % of
Overall Average
1 6,85 105,5
2 6,75 103,9
3 6,61 101,8
4 6,55 100,9
5 6,11 94,1
6 6,03 92,8
Overall Average 6,49
7.5 Analysis of colour shifts
Attempts to generate correlation coefficients for colour shifts were less successful than the efforts for
colour fade. A minimum shift in colour balance of 5 % was used as a threshold for inclusion in this
data set. However, ink sets are often formulated with a mind to minimize any colour shifts, so several
of the ink sets tested were so well balanced that none of the patches on their targets showed a 5 %
colour shift. For example, for Mortsel, Belgium, the least aggressive climate in this study, only 8,1 %
of the possible data points were available. A greater concern is the life cycle of a colour shift during a
weathering test. The colour balance shift is zero both initially and when the colourants have all faded to
white after severe degradation. At some point in between, the colour balance shift reaches a maximum.
For the accelerated data, which is taken frequently, it is easy to see which side of the maximum the data
is at. For one year outdoor data, on the other hand, it is very hard to determine whether a data point
is before or after the maximum colour shift. If it is before the maximum, then a correct data point will
be obtained. If it is after the maximum, then a false data point will be obtained with an acceleration
factor that is higher than the correct data point. This effect skews the data unless one has the ability to
monitor each patch on the outdoor samples and to determine when the maximum colour shift occurs. It
also explains the large standard deviations for the colour shift acceleration factors. Note that the data
in Table 11 has excluded all data points with acceleration factors of 20 or greater, as it is assumed that
those are all false data points on the wrong side of the colour shift maxima. For examples of the colour
shifting behaviour of the materials in this study, see G.4.
Table 11 — Colour shift acceleration factors for 1 y
DSET Chicago Sanary South Milwaukee San Tokyo Mortsel Marly
Arizona France Florida Diego Japan Belgium Switz.
Average Colour 7,62 8,81 8,53 6,46 9,05 7,92 9,54 11,06 11,61
Shift AF
Stdev AF 3,85 3,96 3,98 3,44 4,21 3,68 4,54 4,08 3,02
% of Data Points 18,1 14,4 18,2 27,4 18,0 17,8 13,6 8,1 14,8
Available
There was very little data available for D yellowing. Only three of the materials had white patches
min
that reached a ΔE of 10 in the accelerated test. The apparent acceleration factors for the yellowing were
in the expected range, but it was not a large enough data set be considered representative of the entire
set of 32 materials.
7.6 Two-year data analysis
Another aspect of the round robin test for which sufficient data was lacking was the correlation of two
year outdoor exposure with colour fade. As is implied in Table 3, it is necessary to get 45 – 60 % of the
possible data points before the colour fade acceleration factors converge to their true values. Table 12
shows both 1 y and 2 y data. Especially for the more aggressive climates, not enough data points are in to
consider the 2 y colour fade acceleration factors to be valid. Because of this, the reciprocity performance
of ISO 18930 cannot be determined from this test. The data for the less aggressive climates implies that
any reciprocity failure between 1 y and 2 y of outdoor exposure would be small. It is recommended that
a future test be run with the outdoor exposure extended to 5 y to 10 y in order to evaluate performance
in regard to reciprocity.
Table 12 — Colour fade acceleration factors (AF) by site – 1 y and 2 y
Arizona Chicago Sanary, South San Tokyo, Mortsel, Marly,
FRA Florida Milwaukee Diego JP BEL CH
AF Colour
4,84 7,38 6,13 4,13 7,47 5,61 8,28 8,22 8,49
Fade – 1 y
Material
1,91 3,17 2,19 2,53 2,96 2,00 3,38 4,06 3,85
Stdev – 1 y
% Data
Points 50 54 62 56 59 57 49 63 58
Available – 1 y
AF Colour
7,18 8,42 7,43 6,10 8,77 6,59 9,12 10,35 9,84
Fade – 2 y
Material
2,10 2,95 2,57 2,24 3,16 1,91 2,76 4,07 3,48
Stdev – 2 y
% Data
Points 38 50 48 28 52 27 49 48 56
Available – 2 y
10 © ISO 2018 – All rights reserved

7.7 Correlation coefficients and predictive correlations
The Pearson correlation coefficients for colour fade acceleration factors are shown in Table 13. The data
sets compared are
a) hours of outdoor exposure, and
b) hours of accelerated test to reach the same fade ratio.
Each colour patch of each sample panel used for ISO 18930 accelerated testing is represented by a data
point. The overall average correlation coefficient of 0,677 compares well with xenon arc weathering
[9] [10] [14][15]
results from Crump , Klemann , and Bauer . It is also observed that the correlation coefficients
are highest for the sites at the lowest latitudes. Again, these sites also showed lower coefficients of
variation for their acceleration factors.
Table 13 — R-squared Pearson correlation coefficients
R correlation
Outdoor site
coefficient
DSET Arizona 0,736
Chicago 0,636
Sanary, France 0,708
South Florida 0,772
Milwaukee, Wis. 0,637
San Diego 0,723
Tokyo, Japan 0,606
Mortsel, Belgium 0,631
Marly, Switzerland 0,635
Overall Avg. 0,677
In order to predict acceleration factors for local climates, the relationships between annual climatic
parameters and accelerations factors was explored. Ideally, one would like to select annual climatic
parameters that represent the effects of the three key factors that drive degradation: light, water, and
heat. If these parameters correlate strongly enough with the 18930 accelerated test, it would then be
possible simply to look at the relevant climatic data for a site to predict an acceleration factor before
running any tests on printed materials. In order to select the appropriate parameters, a series of
variables for the nine outdoor sites were plotted against both acceleration factor and its reciprocal,
the number of hours in ISO 18930 per year outdoors (see Table 14). Average annual temperature has
the highest correlation coefficient, so it is selected as representative of thermal effects. Annual solar
radiant exposure also correlates well with acceleration factor, so it may be selected as the proxy for
light exposure. The surprise here is how poorly the variables representative of moisture correlate;
annual precipitation has almost no relationship at all with the acceleration factor. The best climatic
parameter related to moisture is average annual dew point, which still has R below 0,5.
When the selected parameters are plotted versus the acceleration factor to establish appropriate
exponents, and then combined together into an equation, the results are excellent. The predictive
equation for hours in ISO 18930 testing per year outdoors shows an R value of 0,89. This exercise also
uncovers a reason why the correlation coefficients are larger for the low-latitude, aggressive climates.
The aggressive climates are characterized primarily by radiant exposure and temperature. They can
be very humid, like South Florida, or very dry, like Arizona. ISO 18930, and most other accelerated
weathering test cycles, increases the dosage of light and heat more than that of water. So, low-latitude
test sites are more similar to ISO 18930 test conditions and correlate more strongly with it than high-
latitude sites.
Table 14 — Variable Selection for Predictive Correlations
Dependent variable Independent variable or equation R correlation coefficient
h in ISO 18930/y Outdoors Annual solar radiant exposure (GHI) 0,70
Acceleration Factor Annual solar radiant exposure (GHI) 0,76
h in ISO 18930/y Outdoors Average annual temperature 0,85
Acceleration Factor Average annual temperature 0,80
h in ISO 18930/y Outdoors Annual precipitation 0,01
Acceleration Factor Annual precipitation 0,05
h in ISO 18930/y Outdoors Average relative humidity (RH) 0,17
Acceleration Factor Average relative humidity (RH) 0,25
h in ISO 18930/y Outdoors Average annual dew point (TDP) 0,40
Acceleration Factor Average annual dew point (TDP) 0,32
h in ISO 18930/y Outdoors Latitude 0,67
Acceleration Factor Latitude 0,62
0,973 0,618 0,210
h in ISO 18930/y Outdoors (GHI) (AT) (TDP) 0,89
Figure 2 — Results with predictive equation based upon climatic data
7.8 Example — Degradation of Material H4
Material H4 has been selected as an illustrative example of the degradation during the ISO 18930
accelerated weathering test. The material has been chosen here since its degradation rate is fast enough
to illustrate general trends, but slow enough that sample fade was still taking place at the end of the
5 200 h test. Each laboratory tested two replicates which are numbered sequentially (samples 31 and
32 are from the same laboratory, samples 33 and 34 were provided by a different laboratory…). The fade
curves show geometric similarity and very similar density ratios until some divergence is observed
near the end of the test when degradation has become severe. The pattern of geometric similarity for
fade curves from different test instruments was not specific to material H4, but was a general feature of
the data for all of the materials in the round robin test. To convey the levels of degradation characteristic
of material H4, photographs of t
...