Handbook of LED and SSL Metrology: Günther Leschhorn Richard Young

Handbook of LED
and SSL Metrology

Günther Leschhorn
Richard Young
Günther Leschhorn, Richard Young
Handbook of LED and SSL Metrology
With contributions from

Richard Distl
Prof. Dr. Thomas Nägele
Dr. Thomas Attenberger
Dr. Ðenan Konjhodži ć
Dr. Matthias Höh
Published by
Instrument Systems GmbH
Munich, Germany
I
© / Copyright: 2017 Instrument Systems GmbH, München
Umschlaggestaltung, Layout: abc cross media GmbH, München
Cover Image: © tailex, fotolia.com
All rights reserved. No part of this book may be reproduced or transmitted

in any form or any means, electronic or mechanical, including photocopying,
recording, or any information storage and retrieval system, without permission
in writing from the copyright owner. Requests should be made through
Instrument Systems GmbH.
II
Preface
The first edition of the Handbook of LED Metrology was published in

1999 and maintained its popularity over 16 years. We received extremely
positive feedback from customers and people interested in the field of LED
measurement. The handbook was considered as a helpful introduction to
basic terms and definitions and served as a good guideline to test setups
and methodology for accurate measurements on LEDs. Although only about
40 pages long, it covered the basic principles of optical characterization of
LEDs. The content of this first edition was sufficient at this early stage of the
first wave of the Solid-State Lighting (SSL) revolution.
As time moved on, the SSL revolution continued and demanded a more
comprehensive view on the subject of SSL and LED measurement. This led
to the decision to intensively review and extend the existing manuscript. The
outcome is the work at hand entitled Handbook of LED and SSL Metrology.
The content is a summary of knowledge gained by Instrument Systems over
the last 30 years. A lot of technical advances in the field of SSL measurements
made it into scientific papers or were selected as contributions to proceedings
of international conferences and symposia. As a matter of fact, numerous
people assisted in preparing the scientific content of this handbook.
We want to take the advantage to acknowledge a number of people who
contributed in a special way to the preparation of the manuscript and the
technical content.
Thomas Nägele was one of the authors of the first edition and left us an
excellent basis for this updated second edition.
As an application engineer, Ðenan Konjhodži ć contributed with measurements
and evaluations to numerous chapters. We are very thankful for his
contributions.
Thanks also to Matthias Höh who was deeply involved in the preparation of the
manuscript for chapter 9 on LED measurements in the production line.
We are further thankful to Thomas Attenberger for technical editing of the
entire manuscript. His experience in the field of LED and SSL measurements
was greatly acknowledged.
Thanks to Christine Costa, Melanie Maier and Bei-Bei Chuang from the
marketing team. They did a fantastic job in preparing the figures and
coordinating the layout and print of this handbook.
Last but not least, we are sincerely grateful to Richard Distl. He was not
only one of the authors of the first edition, but inspired and launched the
preparation of this second edition during his time as president and CEO of
Instrument Systems.
The authors
III
IV
Contents
Preface ....................................................................................................III
Contents ...................................................................................................V
1 Introduction................................................................................ 1
2 Terms and Definitions in Photometry, Radiometry and

Colorimetry................................................................................. 2
2.1 Photometric and Radiometric Quantities........................................ 2
2.1.1 Luminous Flux and Radiant Power................................................. 4
2.1.2 Luminous Intensity and Radiant Intensity........................................ 4
2.1.3 Illuminance and Irradiance............................................................. 5
2.1.4 Luminance and Radiance.............................................................. 6
2.2 The Cosine Law........................................................................... 7
2.3 Colorimetry................................................................................... 9
2.3.1 Dominant Wavelength.................................................................. 10
2.3.2 Purity.......................................................................................... 10
2.3.3 Just Noticeable Differences and MacAdam Ellipses...................... 11
2.3.4 Correlated Color Temperature...................................................... 12
2.4 Color Rendering Index................................................................. 13
2.5 Wavelength and Spectrum........................................................... 15
2.5.1 Peak Wavelength λp..................................................................... 15
2.5.2 Spectral Bandwidth (FWHM)........................................................ 16
2.5.3 Center Wavelength λ0.5m............................................................... 16
2.5.4 Centroid Wavelength λ c............................................................... 16
3 Standards and Recommendations Applying to LEDs and

SSL Products............................................................................ 17
3.1 CIE 127-2007.............................................................................. 17
3.2 IES LM-79-08.............................................................................. 18
3.3 CIE S025 and EN 13032-4.......................................................... 18
3.4 IES LM-80-08 and TM-21-11....................................................... 21
3.5 ANSI_NEMA C78.377-2008 and Energy Star®.............................. 22
3.6 IES TM-30-15............................................................................. 23
3.7 Zhaga Books............................................................................... 26
3.8 Laboratory Accreditation and Traceability...................................... 28
4 Basic Properties of LEDs......................................................... 30

4.1 Package Design.......................................................................... 30
4.2 Electrical Properties and Ambient Conditions................................ 32
V
4.3 Thermal Properties of High Power LEDs....................................... 34
4.3.1 Thermal Modeling........................................................................ 34
4.3.2 Active Cooling............................................................................. 36
4.3.3 Testing Methods for HP LEDs...................................................... 37
4.4 Characteristics of White LEDs...................................................... 39
4.4.1 Phosphor-Converted White LEDs................................................. 39
4.4.2 3-Chip and 4-Chip White LEDs.................................................... 40
4.4.3 Color Rendering and Efficacy of White LEDs................................. 41
4.5 Basic Properties of OLEDs........................................................... 42
5 Optical Measuring Instruments............................................... 44

5.1 Photometer Design...................................................................... 44
5.2 Comparison of Photometers and Spectroradiometers.................... 47
5.3 Requirements for a Spectroradiometer......................................... 48
5.3.1 Scanning Spectroradiometer........................................................ 49
5.3.2 Array Spectroradiometer.............................................................. 50
5.3.3 Selection Criteria......................................................................... 52
5.4 Stray Light Correction of Array Spectroradiometers...................... 53
5.5 Calibration of Spectroradiometers................................................ 56
5.5.1 Wavelength Calibration................................................................ 56
5.5.2 Spectral Calibration..................................................................... 57
5.5.3 Absolute Calibration and Verification............................................ 57
5.6 Imaging Photometers and Colorimeters........................................ 58
6 Basic Properties of Integrating Spheres and Goniometers.... 62

6.1 Integrating Spheres..................................................................... 62
6.1.1 Integrating Sphere Design............................................................ 63
6.1.2 Integrating Sphere Theory............................................................ 64
6.1.3 Coatings..................................................................................... 65
6.1.4 Sphere Directional Response....................................................... 67
6.2 Goniometers............................................................................... 68
6.2.1 Coordinate Systems.................................................................... 69
6.2.2 Goniometer Types....................................................................... 69
6.2.3 Burning Position Correction......................................................... 72
7 Optical Characteristics of LEDs, Modules, SSL Lamps

and Luminaires......................................................................... 76
7.1 Luminous Flux and Radiant Power............................................... 76
7.1.1 The Integrating Sphere Method.................................................... 76
7.1.2 The Goniophotometer Method..................................................... 80
7.2 Luminous Intensity and Radiant Intensity...................................... 82
7.3 The “Averaged LED Intensity” Concept........................................ 83
7.4 The “Partial LED Flux” Concept.................................................... 84
VI
7.5 Spatial Radiation Characteristics.................................................. 85
7.6 Uniformity and Glare.................................................................... 90
8 Discussion of Sample Measurements with Error Analysis..... 92

8.1 Effects of the Dynamic Measuring Range...................................... 92
8.2 Influence of Stray Light on White LEDs......................................... 93
8.3 Influence of Bandpass................................................................. 96
8.4 Performance Indices.................................................................... 97
8.5 External Influences...................................................................... 98
8.5.1 Accuracy and Stability of the Current Source................................ 98
8.5.2 Quality of LED Test Sockets......................................................... 99
8.5.3 Precision of Mechanical Setup................................................... 100
8.5.4 Temperature Stabilization Time.................................................. 103
9 LED Measurements in the Production Line........................... 105

9.1 Conditions and Requirements in Production Testing..................... 105
9.2 Process Integration.................................................................... 106
9.3 Reproducibility and Accuracy..................................................... 108
9.3.1 Correction Factors..................................................................... 108
9.3.2 Reproducibility........................................................................... 109
9.3.3 Accuracy................................................................................... 109
9.4 Field Installation of Measurement Equipment.............................. 111
9.4.1 Wafer Level Testing.................................................................... 111
9.4.2 Die Level Testing....................................................................... 114
9.4.3 Package Level Testing............................................................... 115
9.4.4 Module Level Testing................................................................. 116
9.4.5 OLED Testing............................................................................ 117
List of Figures...................................................................................... 118
List of Tables........................................................................................ 126
References........................................................................................... 127
About Instrument Systems.................................................................. 130
The Authors.......................................................................................... 131
VII
1 Introduction
Rapid developments in LEDs over the past decade have created a major
growth market with completely new applications. Full color displays for large
areas only became possible with the introduction of high-intensity blue LEDs,
while High Power white LEDs are now widely used in general lighting and
the automotive industry. These applications have placed increasingly stringent
demands on the optical characterization of LEDs, and Solid-State Lighting
(SSL) lamps, modules and luminaires, which serves as the benchmark for
product quality.
Specific expertise is needed in order to achieve precise and reproducible
results. This handbook discusses the special characteristics of LEDs and
emerging OLEDs. It provides an overview of state-of-the-art measurement
equipment and gives recommendations for obtaining accurate measurement
results. The main goal of this handbook is to give readers who are new to this
subject an introduction into LED metrology. However, it also provides a useful
reference work for more experienced readers.
As an introduction, basic terms and definitions used in photometry,
radiometry and colorimetry are described. This develops into definitions of
quantities and details such as the physical properties specific to LEDs and SSL
products. Later sections describe the test setups and methodology required
for accurate measurements. Possible sources of error arising from interactions
between LEDs and measuring instruments are also discussed. The handbook
concludes with a section devoted to the unique requirements of LED testing in
a production environment.
Readers who are short of time can selectively read individual sections.
However, it is recommended to read the entire handbook to obtain an in-depth
understanding of this discipline.
1 Introduction 1
2 Terms and Definitions in Photometry,
Radiometry and Colorimetry
2.1 Photometric and Radiometric Quantities
This section provides a brief overview of important terms and definitions that are
essential for an in-depth understanding and therefore correct use of measuring
instruments. A distinction is drawn between radiometric quantities describing
physical optical radiation properties, photometric quantities describing the
perception of optical radiation by the human eye and colorimetry relating to
the visual perception of color by human beings.
The relevant quantities reflect different conditions that are important to
people in their everyday lives. For example, a distant traffic light will appear
to get brighter as you approach it, until you see it as a circular disc rather
than a point source. Then as you start to get closer it still seems to be getting
bigger but not brighter. While the traffic light appears to be like a point source,
luminous intensity is the relevant quantity, but at a shorter distance the
luminance of the source is more appropriate. Other quantities of interest are
illuminance (e.g. light falling onto the skin or illuminating an object) and total
luminous flux (the entire light emitted in all directions).
Table 1:
Radiometry Symbol Unit
Important radiometric
and photometric Radiant power Фe W
quantities.
Radiant intensity Ie W sr-1
Irradiance Ee W m-2
Radiance Le W m-2 sr-1
Photometry Symbol Unit
Luminous flux Фv lumen (lm)
Luminous intensity Iv lm sr-1 = candela (cd)
Illuminance Ev lm m-2 = lux (lx)
Luminance Lv cd m-2
Table 1 shows similarities between the units of radiometric quantities

and photopic quantities (see the “W” in radiometric quantities and “lm” in
photometric quantities). Each photometric quantity has its corresponding
radiometric quantity, where the suffix “e” in the symbols represents the
radiometric quantity and “v” the photometric equivalent.
One watt of light at 555 nm corresponds to 683 lumens, fixing the
relationship between the quantities radiant power and luminous flux. This
factor varies with wavelength and the variation is defined by the Commission
Internationale de l´Éclairage (CIE), also referred to by the translation
2
quantities (see the “W” in radiometric quantities and “lm” in photometric quantitie
photometric quantity has its corresponding radiometric quantity, where the suffix
symbols represents the radiometric quantity and “v” the photometric equivalent.
One watt of light at 555 nm corresponds to 683 lumens, fixing the relationship betw
quantities
“International Commission radiant power
on Illumination”, andV(λ)
as the luminous
functionflux.
(see This
Figurefactor
1). varies with wavelength
variation
The V(λ) curve describes theisspectral
definedresponse
by the Commission Internationale
function of the human eye inde l´Éclairage (CIE), also re
the wavelength range by from
the 360
translation “International
nm to 830 Commission
nm1 normalized to 1. This on Illumination”,
curve is as the V() func
Figure 1). The V() curve describes the spectral response function of the human e
used to weight the radiometric quantity that is a function of wavelength λ in
wavelength range from 360 nm to 830 nm1 normalized to 1. This curve is used t
order to obtain its corresponding photometric quantity. If Qe(λ) is a spectral
the radiometric quantity that is a function of wavelength  in order to o
radiant quantity, thecorresponding
value of the corresponding photometric
photometric quantity. quantity
If Qe(λ) Qv is radiant quantity, the val
is a spectral
derived by integration of Qe(λ) as follows:
corresponding photometric quantity Qv is derived by integration of Qe(λ) as follows
830 nm
Qv  K m  Q  V   d
360nm
e
The constant Km = 683 lm W-1 refers to the (physical) radiometric unit of the wat
The constant Km = (physiological) photometric
683 lm W-1 refers unit ofradiometric
to the (physical) the lumen. unit of the
watt and the (physiological) photometric unit of the lumen.
1 Figure 1:
The full range is 360 nm to 830 nm but values are very small at Diagram
the extremes
showing and
the it is often
practical purposes to the useful range of 380 nm to 780 nm. V(λ) curve (human eye
response function).
Version 8.8 Page:
There are four basic radiometric and photometric quantities which are
described in the following sub-chapters.
1
The full range is 360 nm to 830 nm but values are very small at the extremes and it is often limited for
practical purposes to the useful range of 380 nm to 780 nm.
2 Terms and Definitions in Photometry, Radiometry and Colorimetry 3

2.1.1 Luminous Flux and Radiant Power
Figure 2:
Luminous flux and
radiant power geometry.
Light from the source
spreads in all directions.
The flux is the amount
of optical radiation (or
visible light) emitted by
the source.
Handbook of LED and SSL Metrology INST
Light is electromagnetic radiation and thus

Handbook dQa kind
of of energy. Radiant power Φe
eLED and SSL Metrology INSTR
is defined as the energy
Handbook Φ
LED 
dQe ofofoptical radiation
e and emitted by a source per unit
SSL Metrology INSTRUMENT S
dt [W].
time dt. The unit of radiant power is the watt
As radiant power relates to light emitted in all directions,
flux or is even referred to simply as radiant flux (see Figu
dQe
Φe corresponding
dQeThe photometric value is called luminous
Φe  dt
spectral radiant power e () as follows:
dtAs radiant power relates to light emitted in all directions, it
As radiant
As radiant power relates to lightpower
flux or relates
emittedis ineven to
all 830 light emitted
referred
directions,
nm to simply
it is in all
asdirections,
sometimes radiant it (see
calledflux is sometim
Figur
The corresponding The

flux
total radiant flux or is or is
even
photometric
even referred
referred
TheΦvcorresponding
corresponding
to
value isphotometric
spectral
    
toassimply
 K m Φephotometric
simply radiant
called luminous
radiant power
360nm
asflux
value

radiant
 V value
e
flux
(
(see
is
 )
flux (see
called
and
as
Figure 2).
 d is called luminous f
Figure 2).
luminous flux and is o
is obtained
follows:
by integrating spectral radiant
spectral powerpower
radiant Φe (λ)  ) as follows:
ase (follows:
830 nm
Φ  K m  Φe    V    d
830 nm
e  36V
 Φ2.1.2.    d Intensity and Radiant In
v
Φv  K m Luminous
0nm
360nm
2.1.2. Luminous Intensity and Radiant Int

2.1.2 Luminous 2.1.2.
Intensity and Radiant
Luminous Intensity
Intensity and Radiant Intensity
Figure 3:
Luminous intensity
and radiant intensity
geometry. Radiation from
a point source emitted
per unit solid angle in a
given direction. Figure 3: Luminous intensity and radiant intensity geometry. R
per unit solid angle in a given direction.
Radiant intensity Ie is defined as radiant

Radiant power dΦe emitted
intensity Ie is per unit solid
defined as angle
radiantdΩpower de em
Figure 3: Luminous intensity and radiant intensity geometry. R
given
in a given direction (refer to Figureper
3). It is direction
expressed (refer
in to
watts Figure
per 3).
steradian
unit solid angle in a given direction. It is
[W expressed
sr -1
]. in watts
Figure 3: Luminous intensity and radiant intensity geometry. Radiation from
per unit solid angle in a given direction.
dΦ
I 
e
e
Radiant intensity I is defined as radiant power d emi
dΩ e e
Radiant intensity Ie is defined
given direction (referas
to radiant power
Figure 3). de emittedin per
It is expressed un
watts
given direction (refer to Figure 3). It is expressed in watts per steradia
dΦe with an active area dA positioned at distanc
A detector
dΦe Iradiant
e  flux de. This configuration assumes a point sou
4 Ie  dΩ
dΩ square law holds true. In this geometry, the distance r a
solid angle d (see also Figure 4).
A detector with an active area dA positioned at distance
Handbook of LED and SSL Metrology INSTRUMENT SYSTEMS G
Handbook
A detector with an active of positioned
area dA LED and SSL Metrology
at distance INSTRUMENT SYSTEMS Gm
r from a light source
measures radiant flux dΦe. This configuration assumes a point source and
dA holds true. In this geometry,
dΩsquare
therefore that the inverse  2 law where dA  r 2 the distance
r the solid angle dΩ (see also Figure 4).
r and the detector area dA define
dA
dΩ  where dA  r 2
r2
Figure 4:
The solid angle dΩ of
a cone is defined as
the ratio of the area dA
cut out on a spherical
surface to the square
of the radius r of the
sphere.
Figure 4: The solid angle dΩ of a cone is defined as the ratio of the area dA cut out on
surface to the square of the radius r of the sphere.
Luminous intensity Iv is obtained from spectral radiant intensity Ie using the

Luminous
Figure intensity
4: The IV isdΩ
solid angle obtained from
of a cone spectral
is defined asradiant intensity
the ratio Ie using
of the area the
dA cut equat
out on a
equation: surface to the square of the radius r of the sphere.
830 nm
I  K intensity
Luminous
v m     
I IV is Vobtained
360nm
e   dfrom spectral radiant intensity Ie using the equati
830 nm
Iv  Km
2.1.3. I    V   and
Illuminance
e d Irradiance
2.1.3 Illuminance and3Irradiance
60nm
Figure 5:
2.1.3. Illuminance and Irradiance Illuminance and
irradiance geometry. A
surface of area dA is
illuminated by a light
source or ambient light.
Handbook of LED and SSL Metrology INSTRUMENT SY
Irradiance Ee is obtained from the ratio of the radiant power de fa

Irradiance Ee is obtained from the ratio of the radiant power dΦe falling onto a
element dA. This quantity is expressed in-2watts per square meter [W m
surface element dA. This
Figure 5: Illuminance
quantity andinirradiance
is expressed watts pergeometry. A surface
square meter [W m of
]: area dA is illuminated by a li
or ambient light.
dΦe
Ee 
Figure 5: IlluminancedA
and irradiance geometry. A surface of area dA is illuminated by a lig
or ambient light.
Version 8.8The following relationship between radiant intensity Ie and irradiance

Page
source is derived from the above formula for irradiance Ee:
dΦe I e  dΩ Ie
Version 2 8.8 Eeand
Terms Definitions  2 and Colorimetry
in Photometry, Radiometry 5
Page
dA dA r
dΦ
element dA. This quantity is expressed in watts per square meter [W
Ee  e
dA
dΦe
Ee 
The following relationship dA
The between
followingradiant Ie and irradiance
intensitybetween
relationship Ee for a I
radiant intensity and irradi
e
point light source is derived
sourcefrom the above
is derived formula
from the above formula E
for irradiance e: irradiance
for Ee:
The following relationship between radiant intensity Ie and irradian
dΦe I e  dΩ I e
Ee 
source 
is derived  2 formula for irradiance Ee:
from the above
dA dA r
dΦe I e  dΩ I e
Ee 
Irradiance can come from

any direction,
 2multiple) sources and from
any (even
dA can dA
Irradiance come r any direction, any (even
from multiple)
any distance (see Figuredistance (see irradiance
5). Although Figure 5).isAlthough
often referredirradiance is often referred t
to in relation
to a lamp, irradiance irradiance
is not is not
actually a actually of
property a property
a source of
but a issource
insteadbuta ismultiple)
instead a
Irradiance can come from any direction, any (even so
given
property of the lightdistance
at surface.
a given surface. When
Whenreferring
referring to tothetheirradiance
irradiance ofofa lamp we ar
(see Figure 5). Although irradiance is often referred to
conditions: the irradiance obtained at a surface when the lamp is
irradiance
a lamp we are implying is not actually
the following a property
conditions: of a source
the irradiance but at
obtained is instead a pro
is arranged at a given distance and orientation relative to the surfa
a surface when thegiven
lamp surface.
is the onlyWhen referring
light source and to isthe irradiance
arranged at a of a lamp we are
given
conditions:
distance and orientationIlluminance to E
relativethe irradiance
v can
the obtained
be calculated
surface and at a surface
from
the optical spectralwhen
axis. irradiance Ee using
the lamp is the
is arranged at a given distance and orientation relative to the surface
Illuminance Ev can be calculated from spectral irradiance Ee using the
830 nm
following formula: Illuminance Ev can be calculated from spectral irradiance Ee using th
Ev  K m
8303nm
 E   V   d
60nm
e
Ev  K m  E   V   d
e
2.1.4.360nmLuminance and Radiance
2.1.4. Luminance and Radiance Flux,

2.1.4 Luminance and Radiance d
Figure 6: Flux,
Luminance and radiance Elemental solid
geometry. Light is d
emitted from a surface of angle d
area dA in a solid angle 
dΩ and a given direction. Elemental solid
angle d

Elemental
surface area, dA
Elemental
Handbook of LED andarea,
surface SSL Metrology
dA INSTRUME
Figure 6: Luminance and radiance geometry. Light is emitted from a su
angle dΩ and a given direction.
Figure 6: Luminance and radiance geometry. Light is emitted from a surfa

Radiance Le is measured Radiance
forand
angle dΩ Le islight
extended
a given measured
sources for
direction. (i.e.extended light
not a point sources (i.e. not a
source)
by the
and is defined by the equation: equation:
Version 8.8 dΦe

Le 
cos   dA  dΩ
Version 8.8
where
where
de represents the radiant flux transmitted by an
dΦe represents the radiant flux transmitted by an elementary beam passing
through a given point and propagating in the so
through a given point and propagating in the solid angle dΩ, containing
given direction;
a given direction;
dA is the area of the section of the beam containing th
 is the angle between the normal to that section and
Radiance is expressed in watts per steradian per square centim
6 the equation for radiance and luminance does not represent
change of flux with solid angle or area) but rather the quotient
element of solid angle and an element of area (see Figure 6).
by the equation:
dΦe
Le 
cosdΦ
  edA  dΩ
Lwhere
e 
dA
cos   dA  dΩ
is the area of the section of the beam containing the given point;
where de represents the radiant flux transmitted by an elementary bea
θ is the angle between the normal to that
through a section and the
given point anddirection of the in the solid angle d, c
propagating
beam. d e represents the
given direction; radiant flux transmitted by an elementary beam
through a given point and propagating in the solid angle d, con
dA given
is the area of the section of the beam containing the given point;
direction;
Radiance is expressed in watts per steradian per square meter [W sr-1 m-2]. Note,
dA and
that the equation for radiance is is
thethe angle
area
luminance between
ofdoes
the sectionthe
of normal
the abeam
not represent to that sectionthe
containing
derivative andgiven
the direction
point; of
-1 -2
(i.e. a rate of change ofRadiance
 withis
flux issolid
expressed
the angle orinarea)
angle betweenwattsbut
perrather
the steradian
normalthe per section
square and
toquotient
that centimeter [W sr ofmth
the direction
of an element of flux the equation
by an element for radiance
of solidinangleand luminance
and an does not
element of square represent a derivative
area centimeter [W sr-1 m-2]. (i.e
Radiance
change is of expressed
flux with solid watts
angleperor steradian
area) but per
rather the quotient of an element oN
(see Figure 6). In strict mathematical
theelement
equation for terms the definition
radiance and could be written as
of solid angle and anluminance
element ofdoes area not
(seerepresent
Figure 6).aInderivative (i.e.
strict mathem
follows: change of flux with solid angle or area)
the definition could be written as follows: but rather the quotient of an element of f
element of solid angle and an element of area (see Figure 6). In strict mathemati
Φ as follows:
the definition could be written
Le  lim e
A , Ω  0 cosΦ
e  A  Ω
Le  lim
A , Ω  0 cos   A  Ω
In practical
In practical measurements, A and measurements, A and
Ω should be small  should
enough be small enough for directional vari
for directional
variations in Φe not Intonot to affect
affect the
practical
the result.
result. Otherwise,
Otherwise,
measurements,
the Φ
the ratio
A and should
ratio
e /(
be
e/(cosθ
cosθ
small ) AΩ) gives
AΩenough the average ra
for directional variati
the
gives the average radiance exact measurement
and the conditions must be specified. Luminance Lv can be
not to affect theexact
result.measurement
Otherwise, the conditions
ratio emust/(cosθbeAΩ) gives the average radi
specified. Luminancethe from
Lv can spectral radiance
be measurement
exact
L e using
calculated fromconditions
spectral the following
radiance
must be
formula:
Le using the Luminance Lv can be c
specified.
following formula: from spectral 830 radiance
nm Le using the following formula:
Lv  K830
m nm Le  λ   V    d
e  λ   V    d
Lv  K m 36L0nm
360nm
2.2. The Cosine Law

2.2 The Cosine2.2.
Law The Cosine Law
Certain quantities, such as radiance and luminance described above, includ
relationship with respect to the viewing angle of the observer or detector.
Certain
Certain quantities, such quantities,
as radiance
certain such
and luminance
descriptions as radiance
described and
inherently above,
imply this luminance
include a described above, include
cosine relationship; for instance “cosin
relationship
cosine relationship with respect to with
the respect
viewing
or “Lambertian emission”. to
angle the
of the viewing
observer angle
or of the observer or detector. M
detector.
certain descriptions
Moreover, certain descriptions inherentlyinherently
imply thisimplycosinethisrelationship;
cosine relationship;
for for instance “cosine
or The cosine relationship
“Lambertian emission”. originates directly from the fact that these quantitie
instance “cosine collector”
planeorin“Lambertian emission”.
their definition. We can define an area within the plane, but when we
The
The cosine relationship
areacosine
from relationship
originates
an the originates
angledirectly from the
apparent directly
size fact from these
that
changes. the fact that
Obviously, whenthese quantities
viewed from n
plane
plane
quantities include a plane in their definition.
( = definition.
in their We
0°) the apparent can define
area an
We can define an
is area area
largest, within
withinand the plane, but
the when viewed from whenwithin
we “
area from an angle the apparent size changes. Obviously, when viewed from norm
( = the
plane, but when we “view” 90°)area
the from
apparent area
an angle theisapparent
zero (you sizecan try looking at a sheet of paper f
changes.
plane ( = 0°) the apparent area is largest, and when viewed from within t
Obviously, when viewed
( = from
90°) normal
the apparent area(θis=zero
to the plane 0 °) the apparent
(you can tryarea is
looking at a sheet of paper fac
largest, and when viewed from within the plane (θ = 90 °) the apparent area is
zero (you can try looking at a sheet of paper face-on and edge-on for a visual
Version
demonstration). At these, 8.8
and all other angles, the apparent area is the actual Pag
area multiplied by the cosine of the angle (see Figure 7).
Version 8.8 Page

Figure 7:
Drawing to illustrate the
cosine law. The apparent
area becomes smaller as
the angle increases with
respect to normal.
In order to correctly measure the irradiance at a plane, the detector must

feature this cosine response so that light from all angles is weighted correctly.
A source of radiance that includes a good cosine distribution is called a
Lambertian source. A perfect diffuse reflector will scatter light so that the
reflection is Lambertian (a cosine distribution) irrespective of the direction of
illumination used.
The response of detectors and the angular distribution of source emission
are often shown on a radial plot. This shows angles around a circle or semi-
circle and intensity or response as the distance from the center, as illustrated
in Figure 8.
Figure 8: Isotropic source Lambertian source 60° LED source
A radial plot showing
examples of sources:
Isotropic (same intensity 0°
in all directions),
Lambertian (the value -30° 30°
changes with the cosine Angle
of the angle) and an LED
with 60 ° view angle.
-60° 60°
-90° 90°
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Intensity
The cosine law therefore represents an ideal behavior, and the quality of
sources, detectors and diffuse reflectors are usually measured by the degree
of deviation from this ideal [1].
8
Lambertian (the value changes with the cosine of the angle) and a LED with 60° view ang
The cosine law therefore represents an ideal behavior, and the quality
detectors
Figure 8: A radial plot showing examples and diffuse
of sources: reflectors
isotropic (same are usually
intensity measured
in all by the degree of deviati
directions),
ideal
Lambertian (the value changes with the
2.3 Colorimetry [1].of the angle) and a LED with 60° view angle.
cosine
The cosine Colorimetry

law therefore
relates to the 2.3.
representsvisual Colorimetry
an perception
ideal behavior, and
of color by the quality eye
the human of sources,
and
detectors
amples and diffuse
of sources: areflectors
isotropic
provides (same are usually
intensity
quantitative and measured
in all by the of
directions),
qualitative description degree
color. of
In deviation
1931 the from
CIE this
ideal
the [1].of the
cosine angle) and a LED with Colorimetry
60° view angle.relates to the visual perception of color by the human eye and
established the X, Y, Z tristimulus system which is based on the assumption
quantitativeofand
that every color is a combination qualitative
the three primary description
colors red,ofgreen
color.and
In 1931 the CIE established
2.3. Colorimetry tristimulus system which is based on the assumption that every color is a co
esents an ideal
blue [2]. The X, Y
behavior, , Z tristimulus
and the quality of sources,
values are obtained by integrating the product
the three primary colors red, green and blue [2]. The X, Y, Z tristimulus values a
are usually measured by thepower
of the spectral degree of deviation
distribution from this
of radiation S(λ) and
by integrating the product of the
thethree colorpower
spectral matching
distribution of radiation S() a
Colorimetry functions
relates to (λ),
the visual perception
(λ) andcolor of color
(λ) matching by the human eye and provides a
(see Figurefunctions x   , y   and z 830
9, left) over the 360 nm to  nm Figure
(see 9, left) over the 36
quantitative wavelength
and qualitative description of color. In 1931 the CIE established the X, Y, Z
range. nm wavelength range.
tristimulus system which is based on the assumption that every color is a combination of
the three primary colors red, green and blue830[2]. nm The X, Y, Z tristimulus values830 arenm obtained 83
X  K power
by integrating the product of the spectral
al perception of color by the human eye and m S 
   x 
distribution
provides a
  d
of Y
radiation K
S(  )
m and  three
Sthe  y    d
 Z  Km
color matching functions x   , y   and z 360
 nm
(see Figure 9, left) over the 360 nm to 830
360 nm 36
cription of color. In 1931 the CIE established the X, Y, Z
ednmonwavelength range.
the assumption that every color is a combination of
nm The X, Y, Z tristimulus
en and blue830[2]. values
Since the830are
nm obtained
eye response curves depend on 830the
nm field of view, two sets of col
X  K mpowerS distribution
spectral     x    dof radiation
functions

Y S(Km) and
have been
Sthe three
 y defined
  d by the K m TheS 
Z  CIE. 2° observer
z    dis commonly us

y   and z360
 nm(see Figure 9, left) over sources whereas
the 360 nm tothe
360 nm 83010° observer is more360suitable
nm for color evaluation of ob
large surface area.
Since the830eye
nm response curves depend The well-known
on830the CIE
nm field of chromaticity
view, two sets coordinates x, y and z are then derive
of color matching
Y  K m have
functions
 S been
  y defined
  d by chromaticity
K m TheS 2°
tristimulus
Z CIE.
the
   observer
values z    dis commonly used for
(X, Y and Z) by normalizing to light
the sum X+Y+Z. As z
sources whereas is
the 10° observer is more360suitable uniquely represented by just x and y
for color evaluation of objects with acoordinates. Plotting value
360 nm nm
large surface area. gives the distinctive shoe shape for monochromatic wavelengths (the mo
Since CIE
The well-known the eye response coordinates
curves depend x, yonandthe zfield
areofthen
view,derived
two setsfromof the
depend on the field of chromaticity
view, two sets of color matching
tristimulus values
color (X, Y
matchingand Z)
functionsby normalizing
have been
y the CIE. The 2° observer is commonly used for light to the
defined bysum
the X+Y+Z.
CIE. The As
2 ° z = 1
observer - is(x+y),
chromaticity is uniquely
commonly represented
used for light by just
sources x and
whereas
ver is more suitable for color evaluation of objects with a y coordinates.
the 10 ° Plotting
observer is values
more of y
suitable and x
gives the distinctive shoe shape
for color evaluation for monochromatic
Version
of objects wavelengths
8.8a large surface
with area. (the monochromatic Pa
ity coordinates x,They well-known
and z areCIE chromaticity
then the x, y and z are then derived
coordinates
derived from
by normalizing
fromtothe
thetristimulus
sum X+Y+Z. (X, z
values As Y= and1 Z- ) (x+y),
by normalizing to the sum X+Y+Z.
nted by just x and y 1-(
As z = coordinates. Plottingis values
x+y), chromaticity uniquelyofrepresented
y and x by just x and y coordinates.
peVersion Handbook
for monochromatic
8.8 Plotting of LED andgives
wavelengths
values
SSL
of y and x (the
Metrology
monochromatic INSTRUMENT
the distinctive shoe shape for monochromatic
SYSTEMS GmbH
Page: 21
wavelengths (the monochromatic locus). All real sources must be combinations
of one or more monochromatic components, so all must lie within the area
locus).byAllthe
bounded real sources mustlocus.
monochromatic be combinations of one or more monochromatic components,
so all must lie within the area bounded
Page: 21 by the monochromatic locus.
X Y Z
x y z  1  ( x  y)
X Y  Z X Y  Z X Y  Z
The right side of Figure 9 shows this chromaticity space according to CIE 1931 for the 2°
Theobserver.
right sideThere are other
of Figure 9 showschromaticity spaces,space
this chromaticity e.g. more uniform
according to chromaticity
CIE scales (CIE
  
1931 for uthe
1960 2 ° observer.
v and u v )are
CIE 1976 There or Lother
a b chromaticity spaces, e.g.
that can be calculated transformation of the x, y
by more
uniform z values. The
and chromaticity scales
CIE(CIE
1960 u vu chromaticity
1960 v and CIE 1976spaceu′ v′ or L*a*b*
is ) for example used to calculate
thatcorrelated color temperature
can be calculated by transformation the x, y2.3.4
(see alsoofSection z values.
and on page 24). The CIE
1960 u v chromaticity space is for example used to calculate correlated color
temperature (see also Section 2.3.4 on page 12).

Figure 9:
The CIE 2 ° observer
color matching
functions (left). 1931 CIE
chromaticity diagram for
2 ° observer (right).
2.3.1 Dominant Wavelength
The dominant wavelength λdom is determined from the chromaticity coordinates

of the measured spectrum. A straight line is taken through the color coordinates
of a reference illuminant and the measured chromaticity coordinates F in
the chromaticity diagram (see Figure 9 right side). The equal energy point
E with chromaticity coordinates x = 0.333 and y = 0.333 is generally taken
as the reference illuminant. The intersection between the straight line and
the boundary of the color diagram (i.e. the monochromatic locus) gives the
dominant wavelength. It is a measure of the color sensation (hue) produced in
the human eye by the light source.
The straight line connecting the end points of the monochromatic locus is
called the purple line. Points on the purple line do not correspond to specific
wavelengths of monochromatic light. Hence, a dominant wavelength is not
defined. Instead of a dominant wavelength, a complementary wavelength can
be assigned. SubtractionHandbook of LED and SSL
of the complementary Metrology
wavelength from white light INSTRUME
yields the color on the purple line.
2.3.2 Purity 2.3.2. Purity
Purity, Pe, is defined from Purity,

CIE 1931Pe, xisydefined
chromaticity
from coordinates
CIE 1931 x,yas:
2
chromaticity coordinates
y F  y 0 x F  x0
Pe  
y d  y 0 x d  x0
where the suffix 0 indicates the white reference point (usually th
where the suffix 0 indicates the test
is the whitesource,
reference
d ispoint (usually thewavelength
the dominant equal energy
intersection.
point E), F is the test source, d is the dominant wavelength intersection.
Most single color LEDs are narrow wavelength band radiator
90% and 100%, i.e. their color cannot be distinguished from a m
2
other color spaces may yield slightly different values
Purity is a measure of colorfulness, known as chroma. Colors
desaturated and those close to the monochromatic locus are sa
10 2.3.3. Just Noticeable Differences and MacAda

Most single color LEDs are narrow wavelength band radiators with a purity
of between 90 % and 100 %, i.e. their color cannot be distinguished from a
monochromatic beam.
Purity is a measure of colorfulness, known as chroma. Colors close to the
white point are desaturated and those close to the monochromatic locus are
saturated.
2.3.3 Just Noticeable Differences and MacAdam Ellipses

Figure 10:
MacAdam ellipses
shown at 10x actual size.
The CIE 1931 x y diagram shows color, but it is apparent from the work of
MacAdam [3] that this color space is not uniform. If, for example, a stimulus of
a certain chromaticity has monochromatic light at some wavelength added to
it, how much does the chromaticity need to change before humans can see it
as a different color? This is the essence of a just noticeable difference (JND)
test. MacAdam found that the shape of a JND around the test chromaticity
formed an ellipse. Figure 10 shows that the size, shape and orientation of the
ellipse changed with color. Uniform color spaces (u v and u′ v′ ) were later
introduced in order to make these ellipses more circular.
Results from MacAdam were used for specific examples. For general
“MacAdam ellipses” at other chromaticities these are taken as ΔE = 1 in
L*a*b* space [4] and transformed to other spaces.

spaces.
2.3.4. Correlated Color Temperature
An important property for white light sources is the correlated colo

2.3.4 Correlated Color Temperature
expressed in Kelvin. Table 2 shows general classifications of CCT
examples.
An important property for white light sources is the correlated color temperature
(CCT) expressed in Kelvin. Table 2 shows general classifications of CCT and
Description
provides typical examples. CCT Example
Table 2:
Warm white approx. 2700 K Incandescent lamp
Description CCT Example
General classification
of correlated color Warm white Neutralapprox.
white 2700 K 3000 to 3500 K
Incandescent lamp Halogen lamp
temperature and
Neutral white 3000 to 3500 K Halogen lamp
examples. Cool white 4100 to 5000 K Compact fluorescent
Cool white 4100 to 5000 K Compact fluorescent lamp
Table 2: General classification of correlated color temperature and examples.
As a blackbody heats up it goes through red, orange, yellow, warm white
3
and cool white. At each temperature the blackbody has specific chromaticity
As a blackbody3 heats up it goes through red, orange, yellow, warm
coordinates and the At each temperature coordinates
line formed by these the blackbody is called
has the Planckian
specific chromaticity coo
locus. Conversely, ifformed by these coordinates is called the Planckian locus. Converse
the chromaticity of a blackbody is known the temperature
can be determined of a is
– this blackbody is known the temperature can be determined
the color temperature.
temperature.
Most sources are not blackbodies and hence may not lie on the Planckian
Most
locus. The correlated sources
color are notrelates
temperature blackbodies and hence
to the blackbody may to
nearest not lie on the
the chromaticity of correlated
the source color
when temperature
expressed in relates
CIE 1960 to uthe blackbody
v space. This nearest to th
source
is the only CIE space wherewhen expressed
the iso-CCT linesin(lines
CIE 1960 u vcorrelated
of equal space. This is the only CIE
color
CCT lines (lines of equal correlated color temperature) are perpendi
temperature) are perpendicular to the Planckian locus.
locus.
4x 6y
u v
12 y  2 x  3 12 y  2 x  3
The correlated color temperature of a source can therefore be ca
The correlated colorchromaticity
temperature coordinates
of a source can by therefore be calculated
finding the fromof the blackb
temperature
the u v chromaticityillustrated in Figure 11. As all blackbodies lie on the Planckian locus
coordinates by finding the temperature of the blackbody
same CCT
closest to it, as illustrated lie on
in Figure 11.a As
lineallatblackbodies
a right angle
lie to
onthe
thelocus.
Planckian
locus, all sources with the same CCT lie on a line at a right angle to the locus.
Figure 11:
CIE 1960 u v diagram,
showing a test source 3
chromaticity and A blackbody is an ideal radiator, also called Planckian radiator, i.e. the emission i
the corresponding Law of radiation. Glowing metal like a filament can be described as a blackbody so
chromaticity on the
Planckian locus joined by
the iso-temperature line
(Fit). The temperature
of the blackbody on Version 8.8
the Planckian locus is
the correlated color
temperature. Duv is the
distance between the
chromaticities.
3
A blackbody is an ideal radiator, also called Planckian radiator, i.e. the emission is described by the
Planck Law of radiation. Glowing metal like a filament can be described as a blackbody source to a
certain extent.
12
Although an iso-temperature line may extend to greens and purples, a green
or purple blackbody does not exist. Care must be exercised when interpreting
CCT values that lie far from the Planckian locus to ensure conclusions are
valid. It is not uncommon to use the value of Duv, the distance from the
Planckian locus, in order to ensure that valid CCT values are obtained.
In addition, the Planckian locus ends in the middle of the CIE u v diagram,
corresponding to the chromaticity of a blackbody at infinite temperature.
The iso-temperature line from this point represents a limit of CCT and this
excludes much of the blue region of the diagram. It is possible for LEDs and
other sources to have chromaticities in this region so they would not have an
equivalent CCT. Example spectra for typical sources with their respective CCT
are shown in Figure 12.
Figure 12:
Examples of source
spectra with their
respective CCT.
2.4 Color Rendering Index
When a light source is used to illuminate objects, the colors of the objects
depend on the spectral distribution of the source. Two sources with the same
chromaticity but different spectral distributions will not render objects in the
same way. Color rendering index (CRI) provides a value for how well or badly a
test source would render colors compared to a reference source.

Figure 13: TCS01 TCS02 TCS03 TCS04 TCS05 TCS06 TCS07 TCS08
Representation of the
rendering of the 14 tiles Ref.
used in color rendering
index calculations under Test
test and reference ab 5.9 4.4 4.5 7.3 6.8 7.0 4.6 8.8
sources. A white tile is
included for reference. TCS09 TCS10 TCS11 TCS12 TCS13 TCS14 White
Ref.
Test
ab 21.3 10.1 9.8 12.2 5.8 2.6 0.0
Figure 13 shows an example for rendering of Test Color Samples (TCS) by

two sources. CIE Publication 13.3 [5] describes the recommended calculation
of the CRI. Details of the calculation will be omitted here but essentially the
method follows these steps:
• The CCT is calculated. This means that if there is no valid CCT the CRI is
also not valid. The recommendation is for Duv to be less than 0.0054 for
valid results.
• The reference illuminant is then selected to be the same CCT as the test
lamp. For a CCT of less than 5000 K the reference is a blackbody and
for greater temperatures it is a calculated spectral distribution representing
different phases of daylight.
• Human vision adapts to various illumination conditions by shifting colors, so

the next stage is to apply chromatic adaptation corrections.
• After this the colors are represented on a uniform color space known as
W*U*V* and the color differences between the test source and reference
illuminant are then calculated. These differences are scaled so that a value
of 100 represents a perfect match to the reference and the value decreases
as the color rendering deteriorates.
• The first 8 tiles are desaturated colors, whereas tiles 9 to 12 are more
saturated. To some extent, tile 13 simulates Caucasian skin tones and tile 14
is a strong green. Each of the 14 colored tiles4 has a special color rendering
index, R1 to R14. The average of the first 8 tiles is called the general color
rendering index, Ra, and it is this single value that is most used. Since its
publication in 1995, the color rendering index has become synonymous
with the general color rendering index, Ra, unless specified otherwise.
4
Some non-CIE color samples exist, but are rarely used. Tile 15, for example, simulates Asian skin tones.
14
For normal lighting a CRI of at least 80 is generally acceptable, but high quality
applications require a CRI of 90 or more. Special lighting applications may
need values close to 100 for critical rendering purposes. The CRI, however, is
irrelevant for single color LEDs (e.g. wall washers for buildings).
2.5 Wavelength and Spectrum
The spectral power distribution of the optical radiation emitted by single

color LEDs differs in many ways from other radiation sources. It is neither
monochromatic like a laser nor broadband like a tungsten lamp, but rather lies
somewhere between these two extremes. The spectrum of such LEDs has a
specific peak wavelength λp depending on the manufacturing process, where
the spectral bandwidth (FWHM) is typically a few tens of nanometers (see
Figure 14). The spectral parameters of LEDs are described in the following
sub-chapters.
1.0 Figure 14:

The spectral power
distribution of a blue LED
0.8
λ p and relevant spectral
parameters.
Relative Intensity
0.6
0.4 λ c
0.2
0.0
350 400 450 500 550 600
Δλ 0.5
Wavelength / nm
2.5.1 Peak Wavelength λp
The peak wavelength is at the maximum intensity of the spectrum. It is easy to

define and is therefore generally given in LED datasheets. However, the peak
wavelength has little significance for practical purposes, since two LEDs may
well have the same peak wavelength but different color perception.
Modern thinking and emerging recommendations do not advise use of
the peak wavelength except for information purposes. Centroid wavelength is
the quantity of choice when specifying the characteristics of a monochromatic
source.

2.5.2 Spectral Bandwidth (FWHM)
The spectral bandwidth at half intensity Δλ0.5 is calculated from the two
wavelengths λ’0.5 and λ”0.5 on either side of λp: Δλ0.5 = λ’0.5 - λ”0.5, where the
intensity falls to half of its maximum.
2.5.3 Center Wavelength λ0.5m
Handbook
The center wavelength corresponds of LED
to the and SSL
wavelength Metrology
halfway between the
half-wavelengths λ’0.5 and λ”0.5.
2.5.4.λc Centroid Wavelength

2.5.4 Centroid Wavelength c
The centroid wavelength λc The centroid

corresponds wavelength
to the c corresponds
“center of gravity” of the plot in to the “cen
Hence it is the wavelength that divides
Figure 14. Hence it is the wavelength that divides the area below the spectrumthe area bel
parts according to the following formula:
graph into two equal parts according to the following formula:
2
   S    d
1
c  2
1
 S    d
The centroid wavelength is ideal for characterizing the radiometric properties

of LEDs (e.g. ultraviolet andThe centroid
infrared LEDs). wavelength is ideal for characterizing th
ultraviolet and infrared LEDs).
3. Standards and Recommenda

and SSL Products
The CIE is the main organization providing doc
measurement of LEDs. Standards published by C
organizations such as ISO (International Orga
(Deutsches Institut für Normung e.V.), IEC (Intern
JSA (Japanese Standards Association), ANSI (Amer
The same effective standard (often with minor
translations) may appear by another designation a
other organizations.
Some organizations produce standards or recomme
or for specific applications. These include: IES (I
IESNA (Illumination Engineering Society of North A
16 Testing and Materials), NEMA (National Electric
(Society of Automotive Engineers), SID (Society fo
Electronics Standards Association), FAA (Federal Av
3 Standards and Recommendations
Applying to LEDs and SSL Products
The CIE is the main organization providing document standards for general
optical measurement of LEDs. Standards published by CIE are sometimes
mirrored by other organizations such as ISO (International Organization for
Standardization), DIN (Deutsches Institut für Normung e.V.), IEC (International
Electrotechnical Commission), JSA (Japanese Standards Association), ANSI
(American National Standards Institute), etc. The same effective standard
(often with minor changes, language or convention translations) may appear by
another designation and number when published by these other organizations.
Some organizations produce standards or recommendations that relate
to LED testing in or for specific applications. These include: IES (Illumination
Engineering Society) and IESNA (Illumination Engineering Society of North
America), ASTM (American Society for Testing and Materials), NEMA (National
Electrical Manufacturers Association), SAE (Society of Automotive Engineers),
SID (Society for Information Display), VESA (Video Electronics Standards
Association), FAA (Federal Aviation Administration), etc.
The following sections give an overview of the most important standards
and recommendations for measurement of LEDs and SSL products.
3.1 CIE 127-2007
CIE 127-2007 [6] deals with single packaged LEDs up to 10 mm in diameter.

Larger LEDs, OLEDs and units containing multiple LEDs should be measured
according to other recommendations, and many of these are currently under
discussion but as yet unpublished. In particular, Solid-State Lighting (SSL)
applications have received much recent attention with the publication of IES
LM-79-08, EN 13032-4 and CIE S025.
The CIE publication 127-2007 is probably the best known and most widely
adopted document for LED testing. Although technically a recommendation, it
is recognized as a de facto standard by the industry5. The publication includes
optical measurements that are based on fixed geometries and require specific
equipment:
• Averaged LED intensity
• Partial LED flux
• Total flux
5
CIE Technical Committee 2-46 “CIE/ISO Standards on LED Intensity Measurements”, which was set up
to establish a standard based on CIE Publication 127:2007, was discontinued in 2010 due to the fact
that the CIE Publication was already considered a de facto standard.
3 Standards and Recommendations Applying to LEDs and SSL Products 17

The first two measurement types include “LED” to emphasize that they are
special definitions applying to LEDs only. Total flux is defined in the same way
as for any other source, but with recommendations on the design and size of
integrating spheres that should be used. These measurement methods are
described in detail in Chapter 7.
3.2 IES LM-79-08
LM-79-08 [7] is an approved North American method developed by IES. It is

not an internationally accepted standard. Nevertheless, LM-79-08 was widely
used worldwide and much content was adopted by the first internationally
agreed measurement standard for SSL sources CIE S025 (see Section 3.3).
The method describes procedures and precautions to perform reproducible
measurements on Solid-State Lighting products. These sources are tested
for total flux and, if required, spatial distribution. Sources should be measured
at 25 ± 1 °C ambient in still air such that any mounting fixtures do not add
extra heat sinking. Self-absorption correction is required and the integrating
sphere used for measurement of luminous flux must be large compared to the
source. If spatial distribution is required, a type C goniometer must be used.
IES LM-79-08 shows differences to the more general IES LM-78-07 [8] which
deals with general lighting total luminous flux measurements in an integrating
sphere. A major difference is the inclusion of 2π measurement geometry for
applicable SSL sources.
3.3 CIE S025 and EN 13032-4
As a purely North American standard, LM-79 lacked the coverage of worldwide

accreditation. A number of national documents, such as the draft standard
DIN 5032-9 in Germany, the CQC and GB Standards in China or the JIS Test
Methods in Japan, existed in parallel to the North American standard. Over a
period of many years, standardization committees have been working to close
this gap by creating an international standard. 2013 therefore saw publication
of the European standard prEN 13032-4:2013 which had been developed by
the Working Group WG7 “Photometry” of the Technical Committee CEN/TC
169 “Light and Illumination”. The secretariat of this committee is managed
by the DIN German Standards Organization. The “Photometry” Working
Committee of the Light Metrology Standards Committee (FNL) within the DIN
German Standards Organization was responsible for drawing up the German
national version. Simultaneously and in close cooperation with the WG7
Working Group, the TC2-71 Technical Committee of the CIE was working on
a reference standard with the same content.
18
In 2015, the standard CIE S025:2015 [9] was published. This represents a
milestone in the development of an international standard for the analysis and
presentation of photometric data from lamps, luminaires and modules based
on LEDs. In contrast to LM-79, which does not include LED modules, the
standard encompasses LED modules, LED lamps, LED light engines and LED
luminaires. The only devices not included in this standard are LED packages
and products based on OLEDs. Adoption of CIE S025 as an ISO/CIE/IEC
“Triple Logo” Standard is anticipated.
The measured quantities covered by the standard include measurement of
luminous flux (including partial luminous flux and derived parameters, such as
luminous efficacy), luminous intensity distributions, luminance and colorimetric
quantities, such as chromaticity coordinates, correlated color temperature
(CCT), distance from the Planckian locus (Duv), color rendering indices
and angular color uniformity. Appropriate test setups recommended for all
measured quantities are defined. In the case of luminous flux, for example,
integrating sphere photometers and integrating sphere spectroradiometers
are recommended for modules, lamps, and small luminaires.
Table 3:
Standard test Tolerance interval Applicable for
The standard test
condition conditions and tolerance
Ambient temperature 25.0 °C ±1.2 °C LED lamps/luminaires, intervals of CIE S025.
light engines
Surface temperature Nominal operating ±2.5 °C LED modules
temperature tp
Air movement Stationary air 0 m/s to 0.25 m/s
Test voltage/ Nominal voltage, ±0.4 % for root mean
Test current nominal current square value (RMS)
AC voltage;
±0.2 % for DC voltage
and current
CIE S025 defines uniform standard test conditions (see Table 3), as well as
special requirements and instrumentation (see Table 4). These conditions are
specified for the laboratory, the environment, and the test instruments. Each
standard test condition is subject to a set value and a tolerance condition which
is specified by a tolerance interval (see Figure 15). Since the definition of the
tolerance interval does not take account of measurement uncertainty, different
accuracy characteristics can be accepted for the measuring device. The range
yielded by deduction of the extended calibration uncertainty (twofold standard
deviation) of the instrument being used is known as the acceptance interval.

Table 4:
Requirement
Summary of special
requirements defined Calibration uncertainty for voltmeters and AC: ≤0.2 %
by the CIE S025 ammeters DC: ≤0.1 %
standard for measuring
instruments. Calibration uncertainty and bandwidth of AC ≤0.5 %
power meters bandwidth ≥100 kHz1
Internal impedance voltmeter ≥1 MΩ2
Drift and fluctuation of the voltage supply Within the acceptance interval for test voltage
and test current
Harmonic content and frequency uncertainty of ≤1.5 %3
operating voltage ±0.2 % of the required frequency
AC component for direct-current supply ≤0.5 % (rms)
Electric and photometric stabilization for the LED lamps and luminaires:
device under test ≥30 min and relative difference of maximum
and minimum measured values of the previous
15 minutes <0.5 %
LED modules:
Operating temperature tp achieved and retained
for 15 min in an interval of ±1 °C
Spectral sensitivity photometer V(λ) mismatch index f1′ ≤3 %
Surface of device under test for measurements 4π: ≤2 % of the inside surface of the sphere
with integrating sphere 2π: diameter of the sphere port ≤1/3 of the
sphere diameter
Cosine correction of the detector for Cosine correction index f2 ≤15 %
measurements with integrating sphere
Repeatability for sphere opening/closing ±0.5 %
Stability of the spectral sensitivity of a sphere <0.5 %
between recalibrations
Wavelength range and wavelength uncertainty 380 – 780 nm
for the spectroradiometer ≤0.5 nm (k = 2)
Bandwidth and scanning interval ≤5 nm
spectroradiometer
Angular alignment and resolution angular ±0.5 °
display goniometer ≤0.1 °
Photometric (test) distance for samples with a Beam angle ≥90 °: ≥5xD
maximal luminous dimension D Beam angle ≥60 °: ≥10xD
Narrow angular distribution / steep gradients:
≥15xD
Large non-luminous areas with maximum
distance S: ≥15x(D+S)
Burning position Measurement in specific burning position or
correction to behaviour of the device under test
in the specified burning position (e.g. with the
auxiliary photometer method)4
1
5 kHz or 30 kHz are authorized without high-frequency components
2
An ever higher internal impedance of the measuring instrument is necessary for devices under test with
high impedance
3
≤3 % for power factors >0.9
4
Not necessary for LED modules with temperature regulation
20
Figure 15:
Relationship between
set value, expanded
measurement
uncertainty, tolerance
and acceptance
interval. The definition
of the tolerance interval
does not take account
of measurement
uncertainty.
The measured parameter must be within the acceptance interval for a

standard-compliant measurement. The measuring results can be corrected
to the set value of the tolerance interval in order to reduce the measurement
uncertainty. The special requirements for the test setup may also be corrected
in some cases.
Concerning goniophotometry, a measurement of the source in the burning
position specified by the manufacturer is not mandatory. If the device under
test is not measured in its designed burning position, a correction (e.g. with
the so called auxiliary photometer method) can be applied. This method is
covered in more detail in Section 6.2.3.
3.4 IES LM-80-08 and TM-21-11
LED sources may fail, but as there is no filament, the failure mode is somewhat
different from incandescent sources. They will normally just continue to emit
a lower level of light throughout their life. Catastrophic failures are rare and
attributable to mechanical causes, e.g. stress due to the differential expansion
rates of dies and encapsulant. It is therefore difficult to define the lifetime of
LEDs by failure rate. IES LM-80-08 [10] deals with lumen depreciation of LEDs
and modules. It recommends measuring the lumen output and chromaticity
over a long period of operation. A spectroradiometer is the recommended
equipment for making such measurements. The LEDs or modules are driven
according to the manufacturer’s instructions at three different case temperatures
(55 °C, 85 °C and a manufacturer-selected temperature). Luminous flux and
chromaticity measurements are made at intervals of less than 1,000 hours and
for a total duration of at least 6,000 hours, though more frequent and longer
measurements of up to 10,000 hours are preferred. Results are reported but
LM-80-08 provides no recommendation for estimations of expected lifetime or
lumen output beyond the test period.
IES TM-21-11 [11] provides a measure of usable lifetime for LEDs and
modules. The testing procedure is similar as for LM-80-08 and an average
of at least 20 samples at each temperature is used. An exponential decay of

Handbook of LED and SSL Metrology INSTR
the form Φ(t) = B exp(-αt), where α is the time constant and B is a scaling
constant, is fitted to the last 5,000 hours data (of 6,000 or 10,000 hour tests)
using a least squares method.constant and B“life”
The projected is aover
scaling
whichconstant, is fitted
the lumens are to the last
10,000 hour tests) using a least squares method. The
maintained above the level p [%] is then
lumens are maintained above the level p [%] is then
ln B / p 
Lp 

The lumen maintenance “life” is then expressed as, e.g. L7
The lumen maintenance “life” is then
the expressed
estimated as, up
time e.g.to
L70which
(6k) = 30,000 hours.
the source will emit more
L70(6k) is the estimated time up to which the source will emit
flux on a 6,000 hours testing base.more than 70 %
of its initial luminous flux on a 6,000 hours testing base.
Calculated lifetimes in excess of 6 times test duration shou
Calculated lifetimes in excess of 6 times
> 36,000 test duration should be expressed
hours.
as, e.g. L70(6k) > 36,000 hours.
3.5. ANSI_NEMA C78.377-2008 and Energ

3.5 ANSI_NEMA C78.377-2008 and Energy Star®
Energy Star®, a US-government backed program to lower
provides a set of specifications for lighting components s
Energy Star®, a US-government backed program to lower energy consumption
saving are simultaneously achieved. Although this is not
by lamps, provides a set of important
specifications for lighting
criteria by whichcomponents
LEDs can sobethat
used in applicat
general
performance and energy saving lighting.
are simultaneously achieved. Although this
is not in itself a standard, it provides
Packagingimportant criteria
of lamps by whichtoLEDs
according can be
Energy Star [12] recom
used in applications ranging from exit signstool
educational to general lighting.
to indicate the CCT.
Packaging of lamps according to Energy Star® [12] recommendations
should include an educational tool to indicate the CCT.
Figure 16:
CIE x y chromaticity
diagram showing the 8
Energy Star® nominal
quadrangles defining
color binning. The red
line is the Planckian
locus. 7-step MacAdam
ellipses are shown as a
reference.
Source: ANSI_NEMA_
ANSLG C78.377-2008
Figure 16 shows the Energy Star® binning limits for integral LED lamps. This
principle of color binning is adopted from the American National Standard
ANSI_NEMA C78.377 “Specifications for the Chromaticity of Solid-State
Version 8.8
22
Lighting Products” [13]. The iso-temperature lines (lines where the color of
the source is closest to the blackbody temperature at the intersection with the
Planckian locus) are shown for comparison. As discussed in Section 2.3.4,
correlated color temperature is defined by the intersection with the Planckian
locus of an iso-temperature line that includes the source chromaticity. This
process is only valid close to the Planckian locus. The combination of CCT
limits and acceptable distance from the locus leads to a quadrangle shaped
definition of color binning. ANSI C78.377 defines 8 nominal6 CCT categories
that are used to specify and communicate white light chromaticity information
(see Table 5). The specified quadrangles are shown in Figure 16 together with
7-step MacAdam ellipses as a reference.
Table 5:
Nominal CCT [K] Target CCT and tolerance [K] Target Duv and tolerance Nominal CCT categories
2700 2725 ± 145 0.000 ± 0.006 according to ANSI_NEMA
C78.377-2008.
3000 3045 ± 175 0.000 ± 0.006
3500 3465 ± 245 0.000 ± 0.006
4000 3985 ± 275 0.001 ± 0.006
4500 4503 ± 243 0.001 ± 0.006
5000 5028 ± 283 0.002 ± 0.006
5700 5665 ± 355 0.002 ± 0.006
6500 6530 ± 510 0.003 ± 0.006
Where white LEDs are used in general lighting they should ideally be the desired
class or bin of white with good color rendering properties. As well as the CCT,
color rendering depends on the spectral distribution of light. The quality of
color rendering required by the user is normally task-based but Energy Star®
[14] states that Ra ≥ 80 is generally required. A perfect score is 100 and the
closer to this one gets the better the color rendering properties of the lamp. For
applications involving critical color discrimination it is not unusual to require a
Ra value in excess of 90 or even 95. CRI is considered for a revision [15] by the
CIE. A brief introduction of the calculation procedure is given in Section 2.4.
3.6 IES TM-30-15
CRI has been the industry reference for decades when it comes to measuring
the color quality of light and has a charming simplicity that makes it easy
for the end user to understand and work with. Nevertheless, CRI is a pure
fidelity (magnitude of difference to original image) metric, based on a very
limited set of color samples and therefore has some substantial drawbacks.
The most severe methodological issues are the small sample set, the choice
6
Nominal CCT is a CCT value at 100 K steps that is closest to the target CCT of the product (value that
the product is designed to produce).

of the reference source and the inherent averaging over the test colors which
leads to a loss of information and induces ambiguity. Moreover, CRI does not
convey exact color appearance e.g. saturated colors and red in particular are
not rendered accurately.
Over time, numerous indices have been proposed as a successor for
CRI. The most important are the Color Quality Scale (CQS), Gamut Area
Index (GAI) and Television Lighting Consistency Index (TLCI). Although some
manufacturers of SSL products use and publish these indices, they have not
accomplished international agreement as a successor for CRI up to now.
In 2015, the Illuminating Engineering Society published IES TM-30-15 [16]
as a method for evaluating the color rendition of light sources. Although it is
not finally clear that CIE will follow the IES proposal and make this method an
internationally agreed color metric standard, the chances for a revised version
are quite high. TM-30 is a more accurate fidelity and also a gamut metric with
additional information and graphical representations. It uses a modern and
uniform color space7 and is built on a set of 99 Color Evaluation Samples
(CES) with spectral properties of real, everyday objects (e.g. paints, textiles,
inks). Like CRI, TM-30 uses a combination of daylight and blackbody locus
as reference source. The main difference to CRI is that TM-30 avoids the
discontinuous step at 5000 K and realizes a smoother transition by blending
reference sources in the range of 4500 K to 5500 K (see Figure 17).
Figure 17: 5000K
Smooth transition of
locus 5500K
daylight and blackbody ght Blackb
yli ody loc
locus as reference Da us
source for TM-30-15.
6000K 4500K
5000K
4000K
The main result of a TM-30 calculation is the Fidelity Index Rf and the Gamut
Index Rg. Rf scores from 0 to 100 and can be interpreted as a more accurate
version of CRI8. The value set of the Gamut Index depends on the achieved
Fidelity Index. It scores from 60 to 140 when Rf > 60. Rg < 100 means
decreasing overall saturation and Rg > 100 means increasing overall saturation.
7
Three dimensional CIE CAM02-UCS
8
Rf tends to produce values a little lower than CRI in direct comparison.
24
Figure 18:
Gamut Index versus
Fidelity Index with
approximate limits for
the combination of the
two measures (left).
The shaded areas
are not achievable for
practicable white light
sources and for sources
on the Planckian locus.
Color Vector Graphic
for a typical white light
source (right). Tangential
vectors to the reference
circle indicate hue
shifts. Vectors pointing
inside/outside the
reference circle indicate
decreased/increased
saturation for the specific
color.
The score of a typical white light source (red dot) in Figure 18 left, indicates an
overall desaturation of Rg = 88 with a Fidelity Index of Rf = 73.
One of the basic concepts of TM-30 is the sub-division of the color space
into 16 so-called hue bins in a radial pattern. For each hue bin, an Rf value can
be calculated (see Figure 19), and chroma and hue shifts can be illustrated
graphically by a Color Vector Graph (see Figure 18 right side). The resulting
vectors for each hue bin can be easily interpreted when compared to the
reference. The whole graph is scaled in a way that the reference has a circular
shape (black circle in Figure 18 right). Tangential vectors to the reference circle
indicate pure hue shifts. Vectors pointing inside or outside the reference circle
indicate decreased or increased saturation for the specific color, respectively.
A perfect match for the source under investigation would be to exactly hit the
reference circle. For the given example in Figure 18, the simple interpretation
would be: The source tends to decrease saturation for green and red colors,
while it tends to have a pure hue shift for turquois and orange colors.
Figure 19:
Fidelity Index Rf by hue
bin.

3.7 Zhaga Books
Zhaga is an international consortium of the lighting industry, which develops

specifications that enable the interchangeability of LED-based light sources
made by different manufacturers. Zhaga’s members include hundreds of
companies from throughout the global lighting industry. The cooperation
is governed by a consortium agreement that defines rules regarding
confidentiality, intellectual property and decision making. Zhaga’s ultimate goal
is to bring consensus and simplification in applications for general lighting by
establishing clearly defined interface specifications.
The Zhaga specifications, so called “Books”, which are still in development,
describe the interface between LED luminaire and LED light engine (LLE). An
overview of Book 1 to Book 18 gives Table 6 and some examples of socketable,
circular LED modules according to Book 5 are shown in Figure 20.
Table 6:
Book number Description Current status
Overview of the Book 1
to Book 18 published or Book 1 Overview and common information relating to the other Approved
in development by the Books
Zhaga Consortium.
Book 2 Socketable drum-shaped LLE with integrated electronic Approved
control gear (ECG), maximum 70 mm diameter, mainly
used in downlight applications
Book 3 Circular LED modules with 50 mm diameter and separate Approved
ECG, mainly used in spot lighting
Book 4 Rectangular LED modules with separate ECG, for high- Approved
intensity outdoor and industrial applications
Book 5 Socketable, circular LED module with 70 mm diameter Approved
separate ECG
Book 6 Compact, socketable, circular LLE with integrated ECG Approved
Book 7 Rectangular LED modules with separate ECG, for indoor Approved
lighting applications
Book 8 Socketable drum-shaped LLE with integrated ECG, Approved
maximum 95 mm diameter, for downlight applications
Book 9 Ring-shaped LED modules with a 12 mm or 25 mm light- Approved
emitting surface (LES) with separate ECG
Book 10 Circular LED modules with 75 mm diameter and separate Approved
Book 11 Circular LED modules with 35 mm diameter and separate In development
Book 12 Rectangular and square LED chip-on-board modules with Approved
a circular LES with separate ECG
Book 13 LED drivers Approved
Book 14 Socketable linear LLEs with integrated driver In development
Book 15 Modules to fit with lens arrays In development
Book 16 Planar circular LLEs with integrated driver In development
Book 17 Spotlight LLEs with integrated driver In development
Book 18 Connectivity socket In development
26
Figure 20:
Example of socketable,
circular LED modules
designed according to
Zhaga Book 5.
Source: GE Lighting
Infusion LED module
product family
Each book defines at least the following set of interfaces between the LED
light engine and LED luminaire: mechanical, photometric, electrical, thermal,
and control interface.
For characterizing the photometric interface, different measurements
have to be performed which include a goniophotometric analysis and a two
dimensional imaging analysis of luminance for some books. In general, the
measurement of luminous flux, luminous intensity distribution, correlated
color temperature and color rendering index is required. The values of CCT
and CRI are communicated using a three-digit code according to IEC/TR
62732:2012 [17]. A goniophotometric measurement is required to ensure
that the luminous intensity distribution is as close as possible to a Lambertian
intensity distribution. For that purpose thresholds of partial luminous flux in 4
so-called CIE flux zones have to be maintained (see Table 7).
Table 7:
Relative Partial Luminous Flux
Relative partial luminous
CIE flux zone y-angle Lambertian Min. value of Max. value of flux tolerances from
(all C-planes) light source spot light LLE spot light LLE Zhaga Book 3. Ideal
(reference) values of a Lambertian
light source are given for
FC1 0 ° - 41.4 ° 43 % 39 % 56 % reference.
FC2 – FC1 41.4 ° - 60 ° 32 % 31 % 37 %
FC3 – FC2 60 ° - 75.5 ° 18 % 11 % 22 %
FC4 – FC3 75.5 ° - 90 ° 7% 0% 7%
For the evaluation of luminance properties, a circular light-emitting surface

is for example divided into five segments as shown in Figure 21. With the
measurement of the average luminance of the five segments, parameters such
as luminance rotational symmetry, luminance center balance and luminance
uniformity can be calculated. For these parameters, Zhaga states limits and
criteria that may vary from Book to Book.
Figure 21:
Luminance property
evaluation areas
A3 A2 according to Zhaga
Book 3.
1/ 5 x LES category diameter A5 LES category diameter
A4 A1

3.8 Laboratory Accreditation and Traceability
Standards apply not only to LEDs, modules, lamps and luminaires; they can
also apply to the laboratories measuring them. Probably the most basic and
best known of common standards is the ISO 9000 series [18] for quality
management systems. This is a general quality management standard that
applies to all industries, not just laboratories, and does not directly address the
correctness of measurements or appropriateness of methods.
ANSI/NCSL Z540 is an old but continuing American standard [19] that deals
specifically with laboratories calibrating measurement and test equipment. For
most laboratories making optical measurements of LEDs, ISO/IEC 17025 [20]
is the current international standard to be met9. The ISO/IEC 17025 standard
not only ensures a rigorous quality management system, but also addresses
the competence of laboratories to make specific measurements or calibrations.
Accreditation to a particular standard guarantees that the laboratory
conforms to the standard in every respect. Accreditation differs from
certification in that it involves a third party, the accrediting body, attesting to
technical competence within a laboratory in addition to its adherence and
operation under a documented quality system. The accrediting body is itself
regulated by the International Laboratory Accreditation Cooperation (ILAC) so
that adherence is equivalent worldwide.
A laboratory that is accredited to ISO/IEC 17025 must therefore not only
have the correct quality management systems in place to ensure competence
in the calibration or measurement quantity, it must also continue to demonstrate
this competence to the accrediting body in an ongoing series of in-depth
audits. These are essential to provide demonstrative proof of competence by
the laboratory, as well as international confidence in the measurement and
calibration procedures and results.
Traceability is defined as “A property of a measurement result whereby the
result can be related to a reference through a documented unbroken chain of
calibrations, each contributing to the measurement uncertainty”.
In most laboratories, the calibration of instruments or sources requires
physical artefacts (lamps, detectors, resistors, meters, etc.) and these in turn
must be calibrated. Their calibration must be traceable to a National Metrology
Institute (NMI)10 reference by an unbroken chain of calibrations complete
with uncertainties for each measurement stage. A lack of documentation or
uncertainty budget for any part of the chain means the chain is broken and the
artifact is no longer traceable.
9
That standard was also duplicated to national standards like DIN EN 17025.
10
Examples of NMIs can be found in many countries and include NIST (National Institute of Standards and
Technology, USA), NPL (National Physical Laboratory, UK), PTB (Physikalisch-Technische Bundesanstalt,
Germany), NMIJ (National Metrology Institute of Japan, Japan), KRISS (Korea Institute of Standards and
Science, South Korea), NIM (National Institute of Metrology, China), etc.
28
When making a measurement or calibration, the laboratory should use the
unbroken traceability chain of all calibrated artifacts to derive the traceability
and uncertainty of their result value.
Traceability is an essential part of calibration and measurements to ensure
correct results. The longer the traceability chain, i.e. the more measurements
and stages between an artefact used in calibration or measurement and the
original calibration by the NMI, the greater the uncertainty in the result. It is
important therefore to keep the traceability chain as short as possible.

List of Figures
Figure 1: Diagram showing the V(λ) curve (human eye response

function) .................................................................................. 3
Figure 2: Luminous flux and radiant power geometry. Light from the
source spreads in all directions. The flux is the amount of
optical radiation (or visible light) emitted by the source .............. 4
Figure 3: Luminous intensity and radiant intensity geometry. Radiation
from a point source emitted per unit solid angle in a given
direction .................................................................................. 4
Figure 4: The solid angle dΩ of a cone is defined as the ratio of the
area dA cut out on a spherical surface to the square of the
radius r of the sphere............................................................... 5
Figure 5: Illuminance and irradiance geometry. A surface of area dA is
illuminated by a light source or ambient light............................. 5
Figure 6: Luminance and radiance geometry. Light is emitted from a
surface of area dA in a solid angle dΩ and a given direction...... 6
Figure 7: Drawing to illustrate the cosine law. The apparent area
becomes smaller as the angle increases with respect
to normal................................................................................. 8
Figure 8: A radial plot showing examples of sources: Isotropic (same
intensity in all directions), Lambertian (the value changes with
the cosine of the angle) and an LED with 60 ° view angle .......... 8
Figure 9: The CIE 2 ° observer color matching functions (left).
1931 CIE chromaticity diagram for 2 ° observer (right)............. 10
Figure 10: MacAdam ellipses shown at 10x actual size............................ 11
Figure 11: CIE 1960 u v diagram, showing a test source chromaticity
and the corresponding chromaticity on the Planckian locus
joined by the iso-temperature line (Fit). The temperature of
the blackbody on the Planckian locus is the correlated color
temperature. Duv is the distance between the chromaticities.... 12
Figure 12: Examples of source spectra with their respective CCT ............ 13
Figure 13: Representation of the rendering of the 14 tiles used in color
rendering index calculations under test and reference sources.
A white tile is included for reference........................................ 14
Figure 14: The spectral power distribution of a blue LED and relevant
spectral parameters............................................................... 15
Figure 15: Relationship between set value, expanded measurement
uncertainty, tolerance and acceptance interval. The definition
of the tolerance interval does not take account of
measurement uncertainty....................................................... 21
118
Figure 16: CIE x y chromaticity diagram showing the 8 Energy Star®
nominal quadrangles defining color binning. The red line is the
Planckian locus. 7-step MacAdam ellipses are shown as a
reference.
Source: ANSI_NEMA_ANSLG C78.377-2008 ......................... 22
Figure 17: Smooth transition of daylight and blackbody locus as
reference source for TM-30-15............................................... 24
Figure 18: Gamut Index versus Fidelity Index with approximate limits for
the combination of the two measures (left). The shaded areas
are not achievable for practicable white light sources and
for sources on the Planckian locus. Color Vector Graphic for
a typical white light source (right). Tangential vectors to the
reference circle indicate hue shifts. Vectors pointing inside/
outside the reference circle indicate decreased/increased
saturation for the specific color............................................... 25
Figure 19: Fidelity Index Rf by hue bin .................................................... 25
Figure 20: Example of socketable, circular LED modules designed
according to Zhaga Book 5.
Source: GE Lighting Infusion LED module product family......... 27
Figure 21: Luminance property evaluation areas according to
Zhaga Book 3........................................................................ 27
List of Figures 119

List of Tables
Table 1: Important radiometric and photometric quantities........................ 2

Table 2: General classification of correlated color temperature
and examples .......................................................................... 12
Table 3: The standard test conditions and tolerance intervals of
CIE S025................................................................................. 19
Table 4: Summary of special requirements defined by the CIE S025
standard for measuring instruments.......................................... 20
Table 5: Nominal CCT categories according to ANSI_NEMA
C78.377-2008......................................................................... 23
Table 6: Overview of the Book 1 to Book 18 published or in
development by the Zhaga Consortium..................................... 26
Table 7: Relative partial luminous flux tolerances from Zhaga Book 3.
Ideal values of a Lambertian light source are given for reference .... 27
126
References
[1] CIE Publication 069-1987, “Methods of Characterizing Illuminance

Meters and Luminance Meters”, 1987.
[2] CIE Publication 015-2004, “Colorimetry, 3rd Edition”, 2004.
[3] G. Wyszecki and W. S. Stiles, “Color Science: Concepts and
Methods, Quantitative Data and Formulae (2nd edition)”, Wiley-
Interscience, 2000.
[4] CIE S 014-4/E:2007, “Colorimetry -Part 4: CIE 1976 L*a*b* Colour
Space”, 2007.
[5] CIE Publication 013.3-1995, “Method of Measuring and Specifying
Color Rendering Properties of Light Sources”, 1995.
[6] CIE 127-2007, “Measurement of LEDs”, 2007.
[7] IES LM-79-08, “Approved Method: Electrical and Photometric
Measurements of Solid-State Lighting Products”, 2008.
[8] IESNA LM-78-07, “IESNA Approved Method for Total Luminous Flux
Measurement of Lamps Using an Integrating Sphere Photometer”,
2007.
[9] CIE S025:2015, “Test Method for LED Lamps, LED Luminaires and
LED Modules”, 2015.
[10] IES LM-80-08, “Approved Method: Measuring Lumen Maintenance of
LED Light Sources”, 2008.
[11] IES TM-21-11, “Projecting Long Term Lumen Maintenance of LED
Light Sources”, 2011.
[12] ENERGY STAR® Program Requirements Product Specification for
Lamps (Light Bulbs). Eligibility Criteria Version 1.0, DRAFT 1, 2011.
[13] ANSI_NEMA_ANSLG C78.377-2008, American National Standard
for electric lamps “Specifications for the Chromaticity of Solid-State
Lighting Products”, 2008.
[14] ENERGY STAR® Program Requirements for Integral LED Lamps -
Partner Commitments, amended 3/22/2010.
[15] CIE TC 1-69 “Color Rendition by White Light Sources”.
[16] IES TM-30-15 “IES Method for Evaluating Light Source Color
Rendition”, 2015.
[17] IEC/TR 62732 Edition 1.0 2012-01, “Three-digit code for designation
of colour rendering and correlated colour temperature”, 2012.
[18] ISO 9001:2008, “Quality management systems – Requirements”,
International Organization for Standardization, 2008.
[19] ANSI/NCSL Z540.3, “Requirements for the Calibration of Measuring
and Test Equipment”, 2006.
References 127
About Instrument Systems
Founded by Richard Distl in Munich in 1986, Instrument Systems is today

one of the world’s leading manufacturers of high-precision array and scanning
spectrometers as well as complex photometric systems. Our name stands
for premium class, innovative products and outstanding expert knowledge in
optical measurement technology. Specialized sales engineers can be relied
on to provide a solution for even the most demanding measurement tasks,
exactly tailored to the needs of our customers.
For many years Instrument Systems has been establishing global standards
for spectroradiometric measurement in the LED industry. We are involved
in standardization committees and associations such as DIN and CIE, and
cooperate with the leading metrological institutes. Virtually all renowned
companies in the automotive and aviation industry place their trust in our
measurement systems for the qualification of lighting components and
displays in the vehicle interior or cockpit. We place the focus of our product
development on the use of our systems not only in laboratories but also in fast
production tests.
Since 2012 we have been a member of the Konica-Minolta Group and

benefit from an international network, supplemented by our experienced
representatives. As a continuously growing, medium-sized technology
company Instrument Systems stands for customer proximity and the highest
level of reliability in product quality, service and support.
130
The Authors
Dr. Günther Leschhorn received his PhD in Physics during his work with the
Max-Planck Institute of Quantum Optics in Munich, Germany. He specialized
in the field of single molecule spectroscopy and ultra-fast laser interaction.
He was a member of the Munich-Centre of Advanced Photonics, a cluster of
excellence of the German Research Foundation and holds a stipend of the
International Max-Planck Research School of Advanced Photon Science.
In 2011 he served as a post-doc at the University of Freiburg (Germany) and

was heading a group interested in the field of quantum computation.
He joined Instrument Systems as product manager for Spectrometers,

Spectroradiometry and Solid-State Lighting in 2012. Today he is head of the
product management department of Instrument Systems.
Dr. Richard Young’s interest and contributions to light measurement span

over 40 years. Dr. Young was educated in England, where he received a B.A.
honours degree and Ph.D. in Chemistry.
In 1983 he moved from academia to industry when joined Glen Creston Ltd.
in London. In 1994 he relocated to the USA and joined Optronic Laboratories,
Inc. where he was promoted to vice president at the end of 1994.
In 2010 he joined Instrument Systems as Chief Scientist. He is an active

member of many CIE technical committees and chairs committees on array
spectroradiometers and on spectroradiometry. Richard Young recently
retired from his role as Chief Scientist and is now working as a consultant for
Instrument Systems.
131
The Handbook of LED and SSL Metrology
Rapid developments in LEDs over the past decade have created a major growth
market with completely new applications. Full color displays for large areas only
became possible with the introduction of high-intensity blue LEDs, while high-
power white LEDs are becoming widely used in general lighting and the automotive
industry. These applications have placed increasingly stringent demands on the
optical characterization of LEDs and Solid-State Lighting devices.
Specific expertise is needed in order to achieve precise and reproducible results.
This handbook discusses the special characteristics of LEDs and emerging
OLEDs. It provides an overview of state-of-the-art measurement equipment and
gives recommendations for obtaining accurate measurement results. The main
goal of this handbook is to give readers new to this subject an introduction into
LED metrology. However, this handbook is also a useful reference work for more
experienced readers.
The “Handbook of LED and SSL Metrology” is a truly exciting work in that it crosses the need of
a broad set of participants in the LED space. Newcomers to LED lighting can use the book to
quickly develop a knowledge base while experienced industry participants will find ongoing value
in the book as a constant technical reference on their bookshelf.
Maury Wright, Editor-in-Chief, LEDs Magazine
Congratulations on this very helpful handbook. It is a fantastic reference with valuable information
about light measurement technology and gives expert advice on the application.
Klaus Ludwig, Segment Leader Luminaires/Multimedia, TÜV SÜD Product Service
This handbook provides the reader with detailed information on the basics of photometry
demonstrated by hands-on application.
Emre Onur, Editor-in-Chief, LICHT Magazine
In this extensively updated edition of the “Handbook of LED and SSL Metrology”, the authors
stay abreast of changes due to the introduction of LEDs. They explain, in a coherent way, all one
needs to know about metrology in general as well as the specifics of SSL metrology. In short, it
is a book every lighting specialist dealing with measurement should have on his or her bookshelf.
Arno Grabher-Meyer, Editor-in-Chief, LED Professional Magazine

Handbook of LED and SSL Metrology: Günther Leschhorn Richard Young

Uploaded by

Copyright:

Available Formats

Handbook of LED and SSL Metrology: Günther Leschhorn Richard Young

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Handbook of LED and SSL Metrology: Günther Leschhorn Richard Young

Uploaded by

Copyright:

Available Formats

Handbook of LED

and SSL Metrology

Handbook of LED and SSL Metrology

With contributions from

All rights reserved. No part of this book may be reproduced or transmitted

The first edition of the Handbook of LED Metrology was published in

2 Terms and Definitions in Photometry, Radiometry and

3 Standards and Recommendations Applying to LEDs and

4 Basic Properties of LEDs......................................................... 30

5 Optical Measuring Instruments............................................... 44

6 Basic Properties of Integrating Spheres and Goniometers.... 62

7 Optical Characteristics of LEDs, Modules, SSL Lamps

8 Discussion of Sample Measurements with Error Analysis..... 92

9 LED Measurements in the Production Line........................... 105

List of Figures...................................................................................... 118

List of Tables........................................................................................ 126

About Instrument Systems.................................................................. 130

The Authors.......................................................................................... 131

2.1 Photometric and Radiometric Quantities

Table 1 shows similarities between the units of radiometric quantities

Version 8.8 Page:

2 Terms and Definitions in Photometry, Radiometry and Colorimetry 3

Handbook of LED and SSL Metrology INST

Light is electromagnetic radiation and thus

The corresponding The

2.1.2. Luminous Intensity and Radiant Int

Radiant intensity Ie is defined as radiant

Luminous intensity Iv is obtained from spectral radiant intensity Ie using the

Handbook of LED and SSL Metrology INSTRUMENT SY

Irradiance Ee is obtained from the ratio of the radiant power de fa

Version 8.8The following relationship between radiant intensity Ie and irradiance

2.1.4. Luminance and Radiance Flux,

Figure 6: Luminance and radiance geometry. Light is emitted from a surfa

Version 8.8 dΦe

2.2. The Cosine Law

2 Terms and Definitions in Photometry, Radiometry and Colorimetry 7

In order to correctly measure the irradiance at a plane, the detector must

The cosine Colorimetry

2 Terms and Definitions in Photometry, Radiometry and Colorimetry 9

2.3.1 Dominant Wavelength

The dominant wavelength λdom is determined from the chromaticity coordinates

2.3.2 Purity 2.3.2. Purity

Purity, Pe, is defined from Purity,

10 2.3.3. Just Noticeable Differences and MacAda

2.3.3 Just Noticeable Differences and MacAdam Ellipses

2 Terms and Definitions in Photometry, Radiometry and Colorimetry 11

An important property for white light sources is the correlated colo

2.4 Color Rendering Index

2 Terms and Definitions in Photometry, Radiometry and Colorimetry 13

Figure 13 shows an example for rendering of Test Color Samples (TCS) by

• Human vision adapts to various illumination conditions by shifting colors, so

2.5 Wavelength and Spectrum

The spectral power distribution of the optical radiation emitted by single

1.0 Figure 14:

2.5.1 Peak Wavelength λp

The peak wavelength is at the maximum intensity of the spectrum. It is easy to

2 Terms and Definitions in Photometry, Radiometry and Colorimetry 15

2.5.3 Center Wavelength λ0.5m

2.5.4.λc Centroid Wavelength

• Human vision adapts to various illumination conditions by shifting colors, so