BSI Standards PublicationOrganic Light Emitting Diode OLED displays Part 5-3: Measuring methods of image sticking and lifetime... ORGANIC LIGHT EMITTING DIODE OLED DISPLAYS – Part 5-3: M
General
The measuring equipment must adhere to the specified structure outlined in each section, with the system diagrams and operating conditions reflecting these requirements Figure 1 illustrates the arrangement of the measuring system, with further details provided in Clause 5.
OLED display module Display surface
Figure 1 – Measuring system and arrangement
Light measuring device (LMD)
The LMD as defined in IEC 62341-6-1:2009 shall be used Specifically, the accuracy of the LMD at 1 degree of the measurement field angle is recommended as being ≤ ±3%, and with a repeatability ≤ ±0,5%
Standard measuring environmental conditions
According to IEC 62341-6-1:2009, section 5.1, standard environmental conditions must be maintained for measurements Specifically, for image sticking assessments, the temperature should be regulated at 25 °C ± 2 °C; if this is not feasible, a temperature-controlled detector must be utilized Additionally, the stability of the LMD should be less than 1/5 of the intended detection difference levels for both luminance and color.
Standard measuring dark-room condition
The standard measuring dark-room conditions specified in IEC 62341-6-1:2009, 5.2, shall be applied.
Standard setup conditions
General
For the measurement area, the minimum radius for measurement with the distance and aperture angle is explained in Table 1
Table 1 – An example of measuring distance and radius size
Adjustment of OLED display modules
The adjustment of OLED display modules specified in IEC 62341-6-1:2009, 5.3.1, shall be applied.
Starting conditions of measurements
Warm-up time refers to the duration from when the supply voltage is activated until the display measurements indicate a change in luminance of less than a specified threshold.
2 % per minute Repeated measurements shall be taken for at least a period of 15 minutes after starting The luminance variations shall also not exceed 5 % during the total measurement.
Test patterns
The test patterns for display devices, including mobile phones, tablet PCs, monitors, and TVs, are illustrated in Figure 2 For mobile and tablet displays, if the OLED panel size is less than a 10 mm radius at a 500 mm measurement distance with a 2-degree aperture angle of the LMD, the aperture angle should be adjusted to encompass the pattern area as specified in Table 1 If adjusting the aperture angle proves challenging, the measuring distance and angle can be modified to achieve a measuring field exceeding 500 pixels Figure 2a) presents the test pattern for all applications, while Figure 2b) outlines the usage method for monitors and TVs To ensure measurement repeatability, the measuring locations from P0 to P4 for TVs, as depicted in Figure 2c), are established, taking into account the uniformity of the OLED display panels or modules.
IEC 2152/13 IEC 2153/13 a) – The test pattern for display devices except monitors and TVs b) – The test pattern for monitor and TV devices
IEC 215413 c) – Image sticking measuring location
Figure 2 – Test pattern for image sticking
Conditions of measuring equipment
The general conditions in IEC 62341-6-1:2009, 5.3.3.1, shall be applied
6 Measuring methods of image sticking
Purpose
The purpose of this method is to measure the image sticking of OLED display panels or modules.
Measuring method
Measuring equipment
The following equipment defined in IEC 62341-6-1:2009, 6.1.2, shall be used: a) power supplies and signal sources for driving, b) LMD.
Measuring procedure
The OLED display modules shall be set in dark-room conditions for measurement
1) Initial measurements on full screen pattern
To ensure optimal performance of OLED display modules, apply a full white screen driving signal across the entire display while maintaining standard power supply conditions However, for certain applications, it may be necessary to reduce the full screen luminance in accordance with section 7.3.1 of IEC 62341-6-1:2009.
Measure the initial spectral radiance or tristimulus values of white at points P0 to P4, as illustrated in Figure 2c Additionally, the initial spectral radiance or tristimulus values of the primary colors can be measured individually.
2) Image burn-in using test pattern
To test display devices, set the OLED display modules to a 0% luminance level across the entire screen, with peak luminance at the center as illustrated in Figure 2a For monitors and TVs, configure the peak luminance level within a 4% window pattern at the center, ensuring a 15% luminance level against the background For further guidance, refer to Annex C of IEC 62087:2011, and if using a different pattern, it should adhere to Annex A of this document and be reported accordingly.
Maintain the test pattern for the designated duration, taking into account the luminance degradation curve Measurements should be taken every hour for the initial 6 hours, followed by every 24 hours for the next 120 hours, and subsequently every 72 hours until the target time under standard measurement conditions Alternatively, the test pattern can be sustained until the target time in these standard conditions.
3) Measurements on full screen pattern
To evaluate the performance of OLED display modules, apply a full white screen driving signal across the entire display Subsequently, measure the spectral radiance or tristimulus values at the same location as the initial measurement Additionally, report the initial and final spectral radiance or tristimulus values for each individual primary color.
All measurements shall be done at the target time of 400 and 500 hours and shall be reported
In Figure 3, an example of the burn-in image is shown
Figure 3 – An example of the burn-in image
Analysis and report
Analysis
6.3.1.1 Luminance and chromatic deviation method
Image sticking can be characterized by luminance and chromatic deviation
The image sticking of luminance IS(t) for white is calculated as follows:
(1) where t is the specified measurement time; t 0 is the initial measurement time;
L i is the luminance of measurement location from P i
Chromatic deviation ∆u’v’(t) 0 caused by image sticking at P 0 over time for white is calculated as follows:
∆ (2) where t is the specified measurement time; t 0 is the initial measurement time;
(u’(t), v’(t)) is the white chromaticity value at the specified time;
(u’(t 0 ),v’(t 0 )) is the white chromaticity value at the initial time
The average of chromatic deviation ∆u’v’(t) AVG caused by image sticking between different measuring locations from P 1 to P 4 for white is calculated as follows:
∆ ∑ i = u i t u t v i t v t t v u (3) where t is the specified measurement time;
(u’ i (t), v’ i (t)) is the chromaticity coordinates of measuring locations of P i (i = 1, 2, 3, 4)
The value of u' and v' can be calculated from the tristimulus value X, Y, and Z using the following equations:
The analysis of image sticking will utilize the ∆E* ab metric within the three-dimensional CIE 1976 L*a*b* color space, as outlined in CIE 15-2004, following the procedure specified in section 6.2.2 Additionally, other three-dimensional uniform color spaces may be employed and documented in the test report Each color point can be represented on the L*, a*, and b* axes of the CIE L*a*b* color space by referencing the peak white tristimulus value (X n, Y n, Z n) at the measuring location P 0 during the initial time t 0, using the appropriate transformation equations.
1 x x x x f t is the specified measurement time;
The L*a*b* i coordinates represent the CIELAB color measurements at various locations P i (where i = 0, 1, 2, 3, 4) The tristimulus values of the reference white at the initial measuring location P 0 and time t 0 are denoted as (X n, Y n, Z n) The color difference formula, ∆E* ab (t) 0, quantifies the color change due to image sticking at P 0 over time for the white reference.
∆ (6) where t is the specified measurement time; t 0 is the initial measurement time;
L*a*b* 0 is the CIELAB colour coordinates of measuring locations of P 0
Average of colour difference formula ∆ E* ab (t) AVG caused by image sticking between different measuring locations from P 1 to P 4 for white is calculated as follows:
E (7) where t is the specified measurement time;
L*a*b* i is the chromaticity coordinates of measuring locations of P i (i = 1, 2, 3, 4).
Report
The typical value of image sticking should be reported along with a specified time, as detailed in Table 2 Additionally, it is important to include reports for other primary colors, such as red, green, and blue.
Table 2 – An example of typical value time
The estimated duration of image sticking can be determined by comparing the reference luminance ratio, chromatic deviation, and color difference, as illustrated in Table 3.
Table 3 – An example of the image sticking time with reference
0,004 Average of chromatic deviation ∆u’v’(t) AVG
Average of colour difference ∆ E* ab (t) AVG 5
The image sticking can be reported after target time, as shown in Table 4
Table 4 – An example of the image sticking data at target time
Luminance ratio (IS) Chromatic deviation ∆u’v’(t) 0at P 0
Average of chromatic deviation ∆u’v’(t) AVG
Colour difference ∆ E* ab (t) 0 at P 0 Average of colour difference ∆ E* ab (t) AVG
7 Measuring methods of the luminance lifetime
Purpose
This method aims to assess the luminance lifetime of OLED display panels or modules Luminance lifetime refers to the duration needed for the luminance to drop to a specified fraction of its initial value during operation Unless stated otherwise, the half luminance lifetime will be utilized for these lifetime measurements.
Measuring method
Measuring equipment
The following equipment shall be used: a) power supplies and signal sources for driving, b) LMD
Measuring procedure
OLED display panels must be tested under standard measuring conditions, specifically in a dark-room environment for accurate luminance measurement A full white screen driving signal at 100% grey level should be applied, with all power supplies set to standard operational conditions It is important to note that for certain display applications, the full screen luminance may be reduced as specified in section 7.3.1 of IEC 62341-6-1:2009.
Measure the initial luminance and keep the above operating conditions and measure the luminance of the device under test (DUT) at the specified time The specified time can be 1, 2,
The luminance behavior during operation is illustrated in Figure 4, highlighting measurements taken over various timeframes: 5, 10, 20, 50, 100, 200, 500, 1,000, and 2,000 days When assessing luminance lifetime, the use of an acceleration method may be permissible, as detailed in Annex B It is essential to report the acceleration conditions, the acceleration ratio, and the theoretical foundation of the applied method.
R el at iv e l um inan ce ( % )
Figure 4 – An example of luminance behavior in operation for an OLED display panel or module
Estimation of luminance lifetime
Measuring luminance lifetime directly can take an impractically long time, often exceeding tens of thousands of hours of panel operation To address this, extrapolation methods are utilized to reduce the measurement duration Luminance lifetime refers to the degradation of light emission in OLED displays, and these extrapolation methods estimate lifetime using a formula that models degradation over time, relying on an understanding of the degradation phenomenon.
The degradation phenomenon shows exponential degradation as follows [1] 1 :
L() (0)exp 1 / (8) where t is the operating time;
L(t) is the luminance value of the degradation phenomena at time t;
L(0) is the initial luminance value of L(t); a is the constant (relaxation time); n is the acceleration factor
1 Numbers in square brackets refer to the Bibliography
However, in the case of luminance degradation of OLED displays, this formula does not coincide with the observed result Other formulae should be chosen
In Equation (9), there is a linear relation between ln(L(0) /L(t)) and ln(t)
[ln( ( )/ ( ))] / ln() / ln( ) ln L 0 Lt =1 n t −1 n a (9)
Figure 5 – An example of lifetime estimation with the extrapolation method
With the linear relation, the lifetime may be estimated, using the extrapolation method (Figure
To evaluate the accuracy of the degradation equation, it is essential to analyze the drift in the estimated lifetime An appropriate formula will exhibit minimal drift in the estimated value over time, as illustrated in Figure 6.
E st im at ed i fet im e ( h)
Time elapsed for lifetime measurement (h)
Figure 6 – An example of estimated lifetime depending on the time elapsed
Analysis and report
Generally, the lifetime follows the Weibull distribution, and the lifetime can be expressed with statistical parameters which represent the Weibull distribution
Shape factor: 7,10 Scale factor: 27 426,6 Average: 25 674,7 Std dev.: 4 257,51
Figure 7 – An example of Weibull distribution of lifetime
The lifetime of a product should be characterized by its typical value and deviation The typical value can be expressed using the average mean time to failure (MTTF) or the scale factor of the Weibull distribution, while the deviation is represented by the standard deviation or the shape factor of the Weibull distribution Additionally, the lifetime may also be indicated by the value at the lower 10% position (B10) in the Weibull distribution.
Table 5 – Examples of lifetime measurement
Calculation method of equivalent signal level
The purpose of this method is to define the procedure to calculate the equivalent signal level for the image sticking of a TV type
A.2 Determining the equivalent signal level
OLED degradation is not directly proportional to current density, leading to the definition of normalized luminance intensity and equivalent current density, which correlate with OLED degradation This metric can be utilized in a usage model-based method for measuring image sticking The normalized luminance intensity operates within RGB linear space and can be transformed into an equivalent signal level for linear to non-linear conversion Additionally, precise image sticking simulations for specific applications can be conducted by calculating one image based on normalized luminance intensity or equivalent current density from a variety of actual usage images and sources.
A.2.2 Calculation of the normalized luminance intensity
The equivalent current density is calculated using the OLED degradation function The OLED degradation function that is normalized by initial luminance is given empirically by the stretched exponential function:
The initial luminance of the OLED device is represented by L(0), while A, K, and m are fitting coefficients specific to the device The current density, denoted as J, is measured in subpixels, and t refers to the duration of the test The coefficients K and m are derived from measurement data, with m defined as the inverse of the acceleration factor n, expressed as m = 1/n.
Equations (8) and (9) Figure A.1 shows the measured luminance degradations and fitted lines according to Equation (A.1)
Figure A.1 – Measured 10 mA/cm 2 to 80 mA/cm 2 OLED degradation values and corresponding modelled functions with m = 1/1,7
This model effectively captures the degradation of OLEDs as a function of time and current density, as illustrated in Figure A.1 By utilizing this model and assuming additive degradation, we can formulate a degradation function for a subpixel that experiences exposure to varying current densities over time.
D n which is the degradation of a pixel by the temporal sequence of current densities
J 1 J 2 J 3 …J n and N over a time period t f D n is calculated as follows
1) Consider a first degradation of an OLED, resulting from exposure to current density J 1 over a time period t f This degradation is expressed as:
2) Consider an alternate degradation D A of an OLED resulting from current density J 2 over a time period t 2 This degradation is expressed as:
We can now define t 2 , such that it is the time required to make D 1 equal to D A Accordingly t 2 can be calculated from (A.2) and (A.3) and is expressed as: t f
By applying Equation (A.4), the time period \( t_2 \) can be adjusted to reflect the variations in current densities \( J_1 \) and \( J_2 \) Consequently, when an OLED is subjected to \( J_1 \) for an initial duration and \( J_2 \) for a subsequent duration, the resulting degradation can be quantified.
3) Consider the degradation of an OLED over a third time interval t 3 with exposure to current density J 3 Degradation D 2A can then be expressed as:
Again, defining t 3 such that it is the time required to make D 2 and D 2A equal, t 3 is calculated from (A.5) and (A.6) and is expressed as:
Therefore after the third time interval, the degradation D 3 is expressed as:
Thus, after N intervals, D n is expressed as:
D n = exp− eff f (A.9) where J eff is the equivalent current density and is expressed as:
Intensity is directly related to current density, which can be calculated by adjusting intensity based on efficiency, luminance, and the area of the OLED Consequently, we can represent the normalized luminance intensity \( I_{\text{eff}} \) in RGB linear space.
The normalized luminance intensity I eff can be transformed to the equivalent signal level
I’ eff by using the equation specified in IEC 61966-2-1:1999 as follows:
I (A.13) where signal level I’ eff is the normalized value from 0 to 1
Another linear to non-linear conversion can also be used for the normalized luminance intensity to the equivalent signal level transformation
A.2.3 Extraction of the equivalent signal level from the IEC 62087:2011 10-min video loop
The extraction of maximum and minimum equivalent signal levels is illustrated using the IEC 62087:2011, 10-minute video loop With experimental data of \( m = \frac{1}{1.7} \) and the transformation equations from IEC 61966-2-1 for non-linear to linear conversions, the OLED device primaries align with those specified in IEC 61966-2-1, accommodating both RGB and RGBW formats The summarized maximum and minimum equivalent signal levels are presented in Table A.1, with specific values for \( m = \frac{1}{1.7} \), gamma of 2.2, a D65 white point, and an 8-bit signal The maximum and minimum levels can be recalculated using equations (A.10) or (A.11) for different \( m \) values, and (A.12) and (A.13) for varying gamma values Signal level distributions are depicted in Figures A.2 and A.3 For RGB pixel format displays, the test input signals for OLED panels can achieve maximum equivalent signal levels of 165, 160, and 163, and minimum levels of 66, 65, and 65 across the display area In contrast, for OLED panels with more than three primaries (RGBW), the procedure must be divided into multiple steps to accommodate various combinations of equivalent signals.
Procedure A: 108,0,0 as the maximum signal level, 22,0,0 as the minimum signal level
Procedure B: 0,85,0 as the maximum signal level, 0,21,0 as the minimum signal level
Procedure C: 0,0,89, as the maximum signal level,0,0,27 as the minimum signal level
Procedure D: 151,151,151 as the maximum signal level, 66,66,66 as the minimum signal level
Table A.1 – Examples of the maximum and the minimum equivalent signal levels (8 bits)
RGB pixel format RGBW pixel format Min signal level Max signal level Min signal level Max signal level
8-bit signal level (scaled I′ eff from 0 to 255)
Fr ac tion Fr ac tion
8-bit signal level (scaled I′ eff from 0 to 255)
8-bit signal level (scaled I′ eff from 0 to 255)
Figure A.2 – Accumulated colour intensity of IEC 62087:2011 10-min video loop in RGB subpixel format with equivalent signal distribution chart based on the left images, respectively
8-bit signal level (scaled I′ eff from 0 to 255)
8-bit signal level (scaled I′ eff from 0 to 255)
8-bit signal level (scaled I′ eff from 0 to 255)
8-bit signal level (scaled I′ eff from 0 to 255)
Figure A.3 – Accumulated colour intensity of the IEC 62087:2011 10-min video loop in W,
R, G, and B format, with equivalent signal distribution chart based on the left images, respectively
Acceleration test of lifetime measurement
The purpose of this method is to reduce the measuring time
This testing method enhances luminance degradation compared to standard operation while maintaining the same degradation mechanism As a result, the shape of the Weibull distribution remains unchanged Figure B.1 illustrates examples of the Weibull distribution from the accelerated lifetime test.
Figure B.1 – Examples of Weibull distributions of accelerated lifetime test
These results can be summarized as shown in Table B.1 The statistical analysis results are shown in Table B.2
In this analysis, the p values for each hypothesis indicate that, despite the significant differences in scale factors between Lifetime 1 and Lifetime 2, the shape factors remain consistent This consistency validates the acceleration applied in this example.
Table B.1 – Summary of the acceleration test results in Figure B.1
Marks in Figure A.1 Red square Black circle
Table B.2 – Statistical analysis results of the accelerated lifetime test in Figure B.1
Approval of same shape factor and same scale factor χ-sq DF p
Approval of same shape factor χ-sq DF p
Approval of same scale factor χ-sq DF p
Degrees of freedom (DF) represent the number of independent observations in a sample, adjusted for the number of population parameters estimated from the data The p-value quantifies the strength of evidence against a null hypothesis, indicating the probability of observing a test statistic as extreme as the calculated value, S, under the assumption that the null hypothesis is true A p-value lower than the significance level leads to the rejection of the null hypothesis The chi-square (χ-sq) distribution, characterized by k degrees of freedom, describes the distribution of the sum of the squares of k independent standard normal random variables.
The effect of an acceleration test may be expressed with an acceleration factor In the examples in Figure B.1, the acceleration parameter is the luminance level By using Equation
The relationship between varying luminance levels and their corresponding lifetimes has been established Specifically, lifetimes \( t_1 \) and \( t_2 \) are associated with luminance measurements \( L(t_1) \) and \( L(t_2) \) Each degradation step can be represented by the equation \( n a t L(t) \).
Constant a of Equation (B.1) is calculated at t 2, and the degradation at t 1 is calculated by using t 2 and Lt 2 , which are expressed as:
In this case the relationship between luminance levels and lifetime may be expressed with Equation (B.4) n
Lifetime i is the lifetime operated with a luminance level of L i ;
L i is the luminance level of each condition; n is the acceleration factor
If the luminance level is used as an acceleration parameter, the typical range of n would be 1,6 ~ 2,0
[1] FÉRY, C., B RACINE, D VAUFREY, H DOYEUX, AND S CINÀ Physical mechanism responsible for the stretched exponential decay behavior of aging organic light-emitting diodes Appl Phys Lett Nov 2005, 87, (213502)
[2] WEIBULL, W A Statistical Distribution Function of Wide Applicability Journal of Applied