High Precision Spectroradiometer Integrating Sphere System

2022/7/7 High Precision Spectroradiometer Integrating Sphere System for LED Light Measurement - LISUN
 Sales@Lisungroup.com |  +8618117273997 Search...  English 
Home Abou t Us P rodu cts  Appl i ca ti on s  Sta n da rds  Su ccessfu l Ca ses News  E -Ca ta l og Con ta ct Us
Home  Products  High Precision Spectroradiometer Integrating Sphere System
High P recision Spectroradiometer Integrating Sphere System

P roduct Categories
P roduct No: LP CE-2(LMS-9000)
Goniophotometer  Get a Quote

Spectroradiometer 
Your email address will not be published. Required fields are marked *
Integrating Sphere 
LED Test Instruments  Name* Company*
CFL Testing Instruments 
Photometer and Colorimeter  Email* Cell/WhatsApp
EMI and EMC Test Systems 

Message*
Electronic Ballast Tester 
Electrical Safety Tester 


Environmental Test Chamber 
Verify Code Send
Plug and Switch Testing 
AC and DC Power Supply 
Object Color and Glossiness Test  61

1
Mask Produce and Test Machine 
Electronic Components Test 
Description Video Download
Related Applications
LPCE-2 Integrating Sphere Spectroradiometer LED Testing System is for single LEDs and LED lighting products light measurement. LED’s
quality should be tested by checking its photometric, colorimetric and electrical parameters. According to CIE 177 , CIE84 , CIE-13.3 , IES LM-
LM-79 and LM-80 Test Solutions 79-19 , Optical-Engineering-49-3-033602 , COMMISSION DELEGATED REGULATION (EU) 2019/2015 , IESNA LM-63-2 , IES-LM-80 and ANSI-
C78.377 , it recommends to using an array spectroradiometer with an integrating sphere to test SSL products. The LPCE-2 system is applied
with LMS-9000C High Precision CCD Spectroradiometer or LMS-9500C Scientific Grade CCD Spectroradiometer, and A molding integrating
LEDs and Luminaire Test Solutions sphere with holder base. This sphere is more round and the test result is more accruacy than the traditional integrating sphere.
https://www.lisungroup.com/products/led-test-instruments/high-precision-spectroradiometer-integrating-sphere-system.html 1/4
 Sales@Lisungroup.com |  +8618117273997 Measures: Search...  English 
• Colorimetric: Chromaticity coordinates, CCT, Color Ratio, Peak Wavelength, Half Bandwidth, Dominant Wavelength, Color Purity, CRI, CQS,
TM-30 (Rf, Rg), Spectrum Test
• Photometric: Luminous Flux, Luminous Efficiency, Radiant Power, EEI, Energy Efficiency Class, Pupil Flux, Pupil Flux Efficiency, Pupil Factor,
Cirtopic Flux, Plant Growth Lamp PAR and PPF
HID Lamp Test Solutions
• Electrical: Voltage, Current, Power, Power Factor, Displacement Factor, Harmonic
• LED Light Decay Test according to LM-80: Flux VS time, CCT VS time, CRI VS time, Power VS time, Power Factor VS time, Current VS time
and Flux Efficiency VS time.
Fluorescent Lamp Test Solutions

Option: The LPCE-2(LMS-9000C) or LPCE-2(LMS-9500C) work with a IS-1.5MT Constant Temperature Integrating Sphere and TMP-8
Multiplex Temperature Tester can fully meet IES LM-82
High Precision LPCE-2(LMS-9000C) System Configuration (It is suitable for the Middle & Small Manufactory or General Test
Automotive Electronics Test
Lab):
Solutions
High Precision CCD Spectroradiometer (LMS-9000C), Optical Fiber (CFO-1.5M), Digital Power Meter (LS2050B/LS2050C/LS2012), DC Power
Source (DC Series), AC Power Source (LSP-500VARC or LSP-500VARC-Pst), Integrating Sphere (IS-1.5MA and IS-0.3M), Standard Light
Source (SLS-50W and SLS-10W), 19 Inch Cabinet (CASE-19IN). You can download the detail PDF here: LPCE-2 (LMS-9000C) High Precision
Related Standards
CCD Spectroradiometer Integrating Sphere System Brochure
Specification:
IEC International Electrotechnical
• Spectral Wavelength Accuracy: ±0.3nm, Wavelength Reproducibility: ±0.1nm
Commission
• Sample Scanning Steps: ±0.1nm
• Accuracy of Chromaticity Coordinate (Δx, Δy): ±0.002 (under Standard A Lamp)
• Correlated Color Temperature CCT: 1,500K~100,000K, CCT Accuracy: ±0.3%
CIE International Commission on • Color Rendering Index Range: 0~100.0, Accuracy: ±(0.3%rd±0.3)
Illumination
• Luminous Flux Range: 0.01-200,000lm; Photometric linear Accuracy: ±0.5%
• Stray light: <0.015%(600nm) and <0.03%(435nm)
• Integration Time: 0.1~10,000ms
SASO Saudi Arabian Standards • It can measure the temperatures inside and outside of integrating sphere
Organization • Flux testing method: spectrum, photometric and spectrum with photometric revision
• The system includes the auxiliary lamp device and the software includes self-absorption function
• It can measure the temperatures inside and outside of integrating sphere
• Connect with PC via USB cable. The English version software can be run in Win7, Win8, Win10 and Win11 (The driver was Certificated by
BIS Bureau of Indian Standards
Microsoft)
• The LM-79 Photometric, Colorimetric and Electricity report can be exported PDF and LED Optical Maintenance test report can be exported
Excel or PDF
NOM Norma Oficial Mexicana
LISUN Model LMS-9000C LMS-9000CUV-VIS LMS-9000CVIS-NIR
Wavelength 350-800nm 200-800nm 350-1050nm
 Sales@Lisungroup.com |  +8618117273997 Search...  English 

Related Technical Articles Scientific Grade LPCE-2(LMS-9500C) System Configuration (It is suitable for the Big & Middle Manufactory or Third Party
Test Lab):
SSL photometric measurement difference Scientific Grade CCD Spectroradiometer (LMS-9500C), Optical Fiber (CFO-1.5M), Digital Power Meter (LS2050C), DC Power Source (DC
between goniophotometer and integrating Series), AC Power Source (LSP-500VARC or LSP-500VARC-Pst), Integrating Sphere (IS-1.5MA and IS-0.3M), Standard Light Source (SLS-50W
sphere and SLS-10W), 19 Inch Cabinet (CASE-19IN). You can download the detail PDF here: LPCE-2 (LMS-9500C) Scientific Grade CCD
Spectroradiometer Integrating Sphere System Brochure
Agriculture lighting PAR, PPF and PPFD
introduce and relevant test instruments
Specification:
• CCD detector: Hamamatsu TE-cooled (Temp: -10°C ±0.05°C) high sensitivity back-thinned detector
Color and spectrum measurement difference
between goniospectraodiometer and integrating • Spectral wavelength accuracy: ±0.2nm, Resolution: ±0.1nm, Sample scanning steps: ±0.1nm
sphere system • Accuracy of chromaticity coordinate (Δx, Δy): ±0.0015 (under Standard A Lamp)
• Correlated color temperature CCT: 1, 500K~100, 000K, CCT accuracy: ±0.2%
Led test instruments supplier in China • Color rendering index range: 0~100.0, Accuracy: ±(0.3%rd±0.3)
• Photometric linear Accuracy: ±0.2%
Integrating sphere setup for the LED light
• Stray light: <0.015%(600nm) and <0.03% (435nm)
measurement
• Integration time: 0.1ms-60s
What is the Integrating sphere, and how does it • It can measure the temperatures inside and outside of integrating sphere
work • Flux testing method: spectrum, photometric and spectrum with photometric revision
• The system includes the auxiliary lamp device and the software includes self-absorption function
Integrating sphere: Sphere with light source
• It can measure the temperatures inside and outside of integrating sphere
• Connect with PC via USB cable. The English version software can be run in Win7, Win8, Win10 and Win11 (The driver was Certificated by
Based on the LM-79 standard, how to use the
integrating spheres and spectrophotometers to Microsoft)
test the LED luminaire • The LM-79 Photometric, Colorimetric and Electricity report can be exported PDF and LED Optical Maintenance test report can be exported
Excel or PDF
Agricultural plant growth market development
How you can use light source in an integrating LISUN Model LMS-9500C LMS-9500CUV-VIS LMS-9500CVIS-NIR
sphere to get product reliability results
Wavelength 350-800nm 200-800nm 350-1050nm
Related Successful Case
Taejin Choi
French-LISUN engineer repair LSG-1700B
probe for customers free of charge 2020-01-12
Germany-Customers independently installed It was an essential product for manufacturing and research. I purchased it after confirming that it is
LPCE-2 (LMS-9000) High Precision a CIE/IEC certified product. The quality of the product is excellent and the satisfaction with the
Spectroradiometer Integrating Sphere System
price is high. Employees are also satisfied with their response to repairs and after-sales services.
by watching the video
France-Successfully installed LSG-1700B high- Tags：Integrating Sphere , LPCE-2 , LPCE-2 (LMS-9000A) , LPCE-2 (LMS-9000B ) , LPCE-2 (LMS-9000C) , LPCE-2(LMS-9000) ,
precision goniophotometer by watching the
Spectroradiometer
video
Sales@Lisungroup.com
Oman-Successfully
 |  +8618117273997
installed LPCE-2 (LMS- Search...  English 
9000B) high-precision integrating sphere
system and LSG-1800B Goniophotometer by
watching the vedio Related P roducts
Turkey-Install the LSG-1800BCCD High

Precision Rotation Luminaire Goniophotometer
by watching the vedio
Turkey–Install LSG-1700B Goniophotometer by

 
watching the video
Iran-Free installation and training for LPCE-

2(LMS-9000A)High precision spectrometer
integrating sphere system
India-Installation and training for LSG-1700B

Goniophotometer and Maintenance of LPCE-2 IS-*MA Integrating Sphere With CFO-1.5M Optical Fiber SLS-50W Standard Light Source DC3005 Digital CC and CV DC
Integrating Sphere Testing System Holder Base Power Supply
India-LISUN engineer visit India to do

installation and training for LPCE-2 High
Precision Spectroradiometer Integrating Sphere
System
Egypt – Installation of LSG-1800B

Goniophotometer and LPCE-2 (LMS-9000B)
Integrating Sphere System
Goniophotometer | Integrating Sphere | Surge Generator      © Copyright 2003 | Blog | Privacy
Surge Generator | Emc Test System | Emi Receiver | Electrical Safety Tester | Temperature Chamber | Salt Spray Test | Environmental Chamber | Sitemap
ISBN 978 3 901 906 58 9
om
.c
up
ro
u ng
is
.L
w
w
w
MEASUREMENT OF LEDS
CIE 127:2007
2nd edition
UDC: 53.085.342 Descriptor: Light sources, indicating devices
535.24 Photometry
621.3.032.32 Light-emitting elements
--`,,```,,,,````-`-`,,`,,`,`,,`---
Copyright International Commission on Illumination

Provided by IHS under license with CIE
No reproduction or networking permitted without license from IHS Not for Resale
THE INTERNATIONAL COMMISSION ON ILLUMINATION
The International Commission on Illumination (CIE) is an organisation devoted to international co-operation and exchange of
information among its member countries on all matters relating to the art and science of lighting. Its membership consists of
the National Committees in about 40 countries.
The objectives of the CIE are :
1. To provide an international forum for the discussion of all matters relating to the science, technology and art in the
fields of light and lighting and for the interchange of information in these fields between countries.
2. To develop basic standards and procedures of metrology in the fields of light and lighting.
3. To provide guidance in the application of principles and procedures in the development of international and national
standards in the fields of light and lighting.
4. To prepare and publish standards, reports and other publications concerned with all matters relating to the science,
technology and art in the fields of light and lighting.
5. To maintain liaison and technical interaction with other international organisations concerned with matters related to
the science, technology, standardisation and art in the fields of light and lighting.
The work of the CIE is carried on by seven Divisions each with about 20 Technical Committees. This work covers subjects
ranging from fundamental matters to all types of lighting applications. The standards and technical reports developed by
these international Divisions of the CIE are accepted throughout the world.
A plenary session is held every four years at which the work of the Divisions and Technical Committees is reviewed,
reported and plans are made for the future. The CIE is recognised as the authority on all aspects of light and lighting. As
such it occupies an important position among international organisations.
LA COMMISSION INTERNATIONALE DE L'ECLAIRAGE
La Commission Internationale de l'Eclairage (CIE) est une organisation qui se donne pour but la coop ération internationale
et l' échange d'informations entre les Pays membres sur toutes les questions relatives à l'art et à la science de l' éclairage.
om
Elle est compos ée de Comit és Nationaux repr ésentant environ 40 pays.
Les objectifs de la CIE sont :
1. De constituer un centre d' étude international pour toute mati ère relevant de la science, de la technologie et de l'art de
la lumi ère et de l' éclairage et pour l' échange entre pays d'informations dans ces domaines.
.c
2. D' élaborer des normes et des m éthodes de base pour la m étrologie dans les domaines de la lumi ère et de l' éclairage.
3. De donner des directives pour l'application des principes et des m éthodes d' élaboration de normes internationales et
nationales dans les domaines de la lumi ère et de l' éclairage.
up
4. De pr éparer et publier des normes, rapports et autres textes, concernant toutes mati ères relatives à la science, la
technologie et l'art dans les domaines de la lumi ère et de l' éclairage.
5. De maintenir une liaison et une collaboration technique avec les autres organisations internationales concern ées par
ro
des sujets relatifs à la science, la technologie, la normalisation et l'art dans les domaines de la lumi ère et de
l' éclairage.
Les travaux de la CIE sont effectu és par 7 Divisions, ayant chacune environ 20 Comit és Techniques. Les sujets d' études
ng
s' étendent des questions fondamentales, à tous les types d'applications de l' éclairage. Les normes et les rapports
techniques élabor és par ces Divisions Internationales de la CIE sont reconnus dans le monde entier.
Tous les quatre ans, une Session pl éni ère passe en revue le travail des Divisions et des Comit és Techniques, en fait
rapport et établit les projets de travaux pour l'avenir. La CIE est reconnue comme la plus haute autorit é en ce qui concerne
u
tous les aspects de la lumi ère et de l' éclairage. Elle occupe comme telle une position importante parmi les organisations
internationales.
is
.L
DIE INTERNATIONALE BELEUCHTUNGSKOMMISSION
Die Internationale Beleuchtungskommission (CIE) ist eine Organisation, die sich der internationalen Zusammenarbeit und
w
dem Austausch von Informationen zwischen ihren Mitgliedsl?ndern bez üglich der Kunst und Wissenschaft der Lichttechnik
widmet. Die Mitgliedschaft besteht aus den Nationalen Komitees in rund 40 L?ndern.
w
Die Ziele der CIE sind :

1. Ein internationaler Mittelpunkt f ür Diskussionen aller Fragen auf dem Gebiet der Wissenschaft, Technik und Kunst der
w
Lichttechnik und f ür den Informationsaustausch auf diesen Gebieten zwischen den einzelnen L?ndern zu sein.
2. Grundnormen und Verfahren der Me?technik auf dem Gebiet der Lichttechnik zu entwickeln.
3. Richtlinien f ür die Anwendung von Prinzipien und Vorg?ngen in der Entwicklung internationaler und nationaler Normen
auf dem Gebiet der Lichttechnik zu erstellen.
4. Normen, Berichte und andere Publikationen zu erstellen und zu ver?ffentlichen, die alle Fragen auf dem Gebiet der
Wissenschaft, Technik und Kunst der Lichttechnik betreffen.
5. Liaison und technische Zusammenarbeit mit anderen internationalen Organisationen zu unterhalten, die mit Fragen
der Wissenschaft, Technik, Normung und Kunst auf dem Gebiet der Lichttechnik zu tun haben.
Die Arbeit der CIE wird in 7 Divisionen, jede mit etwa 20 Technischen Komitees, geleistet. Diese Arbeit betrifft Gebiete mit
grundlegendem Inhalt bis zu allen Arten der Lichtanwendung. Die Normen und Technischen Berichte, die von diesen
international zusammengesetzten Divisionen ausgearbeitet werden, sind von der ganzen Welt anerkannt.
Tagungen werden alle vier Jahre abgehalten, in der die Arbeiten der Divisionen überpr üft und berichtet und neue Pl?ne f ür
die Zukunft ausgearbeitet werden. Die CIE wird als h?chste Autorit?t f ür alle Aspekte des Lichtes und der Beleuchtung
angesehen. Auf diese Weise unterh?lt sie eine bedeutende Stellung unter den internationalen Organisationen.
Published by the
COMMISSION INTERNATIONALE DE L'ECLAIRAGE
CIE Central Bureau
Kegelgasse 27, A-1030 Vienna, AUSTRIA
Tel: +43(1)714 31 87 0, Fax: +43(1) 714 31 87 18
e-mail: ciecb@cie.co.at
WWW: http://www.cie.co.at/
CIE 2007 – All rights reserved
--`,,```,,,,````-`-`,,`,,`,`,,`---

ISBN 978 3 901 906 58 9
-
--
`
,
,
`
,
`,
,
`,
,
`
-
`
-
``
`
,`
,
,
,
`
`
`,
,
-`
-
om
.c
up
ro
u ng
is
.L
w
w
w
MEASUREMENT OF LEDS
CIE 127:2007
2nd edition
UDC: 53.085.342 Descriptor: Light sources, indicating devices
535.24 Photometry
621.3.032.32 Light-emitting elements

CIE 127:2007
This Technical Report has been prepared by CIE Technical Committee 2-45 of Division 2
"Physical Measurement of Light and Radiation" and has been approved by the Board of
Administration of the Commission Internationale de l'Eclairage for study and application. The
document reports on current knowledge and experience within the specific field of light and
lighting described, and is intended to be used by the CIE membership and other interested
parties. It should be noted, however, that the status of this document is advisory and not
mandatory. The latest CIE proceedings or CIE NEWS should be consulted regarding possible
subsequent amendments.
Ce rapport technique a é t é é labor é par le Comit é Technique CIE 2-45 de la Division 2

"Mesures physiques de la lumi è re et des radiations" et a é t é approuv é par le Bureau de la
Commission Internationale de l'Eclairage, pour é tude et emploi. Le document expose les
connaissances et l'exp é rience actuelles dans le domaine particulier de la lumi è re et de
l' é clairage d é crit ici. Il est destin é à ê tre utilis é par les membres de la CIE et par tous les
int é ress é s. Il faut cependant noter que ce document est indicatif et non obligatoire. Il faut
consulter les plus r é cents comptes rendus de la CIE, ou le CIE NEWS, en ce qui concerne
des amendements nouveaux é ventuels.
om
Dieser Technische Bericht ist vom CIE Technischen Komitee 2-45 der Division 2
"Physikalische Messungen von Licht und Strahlung" ausgearbeitet und vom Vorstand der
Commission Internationale de l'Eclairage gebilligt worden. Das Dokument berichtet ü ber den
.c
derzeitigen Stand des Wissens und Erfahrung in dem behandelten Gebiet von Licht und
Beleuchtung; es ist zur Verwendung durch CIE-Mitglieder und durch andere Interessierte
up
bestimmt. Es sollte jedoch beachtet werden, da? das Dokument eine Empfehlung und keine
Vorschrift ist. Die neuesten CIE-Tagungsberichte oder das CIE NEWS sollten im Hinblick auf
ro
m?gliche sp?tere ?nderungen zu Rate gezogen werden.

ng
Any mention of organisations or products does not imply endorsement by the CIE. Whilst
every care has been taken in the compilation of any lists, up to the time of going to press,
u
these may not be comprehensive.

is
.L
Toute mention d'organisme ou de produit n'implique pas une pr é f é rence de la CIE. Malgr é le
soin apport é à la compilation de tous les documents jusqu' à la mise sous presse, ce travail
w
ne saurait ê tre exhaustif.

w
Die Erw?hnung von Organisationen oder Erzeugnissen bedeutet keine Billigung durch die
w
CIE. Obgleich gro?e Sorgfalt bei der Erstellung von Verzeichnissen bis zum Zeitpunkt der
Drucklegung angewendet wurde, ist es m?glich, da? diese nicht vollst?ndig sind.
CIE 2007 – All rights reserved
II
--`,,```,,,,````-`-`,,`,,`,`,,`---

CIE 127:2007
The following members of the TC 2-45 "Measurement of LEDs" took part in the preparation of
this report. The committee comes under Division 2 "Physical Measurement of Light and
Radiation". This present publication replaces CIE 127-1997 “ Measurement of LEDs ” .
Members:
Goodman, T. UK
Heidel, G. Germany
Muray, K. USA (Chair)
Ohno, Y. USA
Sauter, G. Germany
Schanda, J. Hungary
Steudtner, W. Germany
Young, R. USA
om
Advisors:
.c
Ashdown, I. Canada up
Bando, K. Japan
Distl, R. Germany
ro
Gugg-Helminger, T. Germany
ng
McKee, G. USA
Sapritsky, V. Russia
u
Schutte, J. Germany
is
Sliney, D. USA
.L
Sperling, A. Germany
w
Stolyarevskaya, R. Russia
--
-
Valenti, T. USA
w
`
,,
`
`,
,
,`
,
,
`-
w
`
-`
`
``
,
,
,,
`
``
,
`,
-
-
III
CIE 127:2007
TABLE OF CONTENTS
SUMMARY VI
RESUME VI
ZUSAMMENFASSUNG VI
1. INTRODUCTION 1
-
1.1 Scope 1 -
-
`,
,
,`
1.2 Terminology 1 `
,
,`
,
1.3 Purpose of the report 1 `,

-
-`
`
`
1.4 Categories of LED measurement 2 ``

,
,,
,
`
1.4.1 Laboratory measurements 2 `

`
,
`,
-
1.4.2 Bulk testing 2 -
2. PROPERTIES OF LEDS 3
2.1 Optical properties of LEDs 3
2.1.1 Spatial distribution 3
2.1.2 Spectral distribution 4
om
2.1.3 Area of emittance 5
2.2 Electrical characteristics 5
2.2.1 Electrical operating conditions 5
.c
2.2.2 Operation of reference standards 5
2.2.3 Time dependent operation 6
up
2.2.4 Forward voltage 7
2.2.5 Ambient temperature 8
2.3 Influence of temperature on the radiation 8
ro
2.3.1 Shift of peak wavelength with temperature 8

2.3.2 Effects of temperature on efficiency and efficacy 9
ng
2.4 Production tolerances 9

3. PROPERTIES OF THE PHOTOMETER / RADIOMETER 9
u
3.1 Detectors 10
3.2 Angular and spatial responsivity of photometers / radiometers 10
is
3.3 Spectral responsivity of the photometers / radiometers 10

.L
3.3.1 Photometer to measure white LEDs 11

3.3.2 Photometer to measure coloured (non-white) LEDs 11
w
4. QUANTITIES DEFINING SPATIAL RELATIONS 11

w
4.1 Normalisation factor and relative spatial distribution 11

4.2 Measurement of directional quantities 13
w
4.2.1 Luminous intensity 13

4.2.2 Illuminance 13
4.2.3 Location of the effective emitting surface 14
4.2.4 "Near-field" and "far-field" measurement conditions 14
4.3 Averaged LED Intensity 14
4.4. Measurement of spatial and directional properties 15
5. MEASUREMENT OF AVERAGED LED INTENSITY 16
5.1 Substitution method 16
5.1.1 Substitution with fewer standards 16
5.2 Applying spectral mismatch correction 16
5.3 Use of a spectroradiometer 17
5.4 Detector-referenced method 17
6. MEASUREMENT OF LUMINOUS FLUX 17
6.1 Measured quantities 17
6.1.1 Total luminous flux 18
6.1.2 Partial LED Flux 18
6.2 Methods of flux measurement 19
6.2.1 Goniophotometer method 19
IV
CIE 127:2007
6.2.2 Integrating sphere method 20

6.2.3 Methods for sphere calibration and correction 22
7. SPECTRAL MEASUREMENT 23
7.1 The concept of spectral distribution 23
7.1.1 Spectral concentration 23
7.1.2 Normalisation factor and relative spectral distribution 24
7.2 Quantities related to spectral distribution 24
7.2.1 Peak wavelength 24
7.2.2 Spectral bandwidth at half intensity levels 24
7.2.3 Centre wavelength of half intensity bandwidth 24
7.2.4 Centroid wavelength 25
7.3 Colorimetric quantities determined from the spectral distribution 25
7.3.1 Dominant wavelength 26
7.3.2 Purity 26
7.4 Spectral measurement of LEDs 27
7.4.1 Irradiance mode 27
7.4.2 Total flux mode 28
7.4.3 Partial flux mode 28
7.4.4 Consideration for bandwidth and scanning interval 30
om
7.4.5 Other uncertainty components 30
8. REFERENCES 31
.c
up
ro
u ng
is
.L
w
w
w
--`,,```,,,,````-`-`,,`,,`,`,,`---
V
CIE 127:2007
MEASUREMENT OF LEDS
SUMMARY
This report is an update of the previously published CIE Technical Report CIE 127-1997.
There are significant differences between LEDs and other light sources which made it
necessary for the CIE to introduce new quantities for their characterization with precisely defined
measurement conditions. New quantities introduced here are "Averaged LED Intensity" and
"Partial LED Flux".
The report describes in detail the measurement conditions for ALI (Averaged LED
Intensity), Total and Partial LED Flux and Spectral Power Distribution. It is shown that
measurements by substitution method using LED standards can be simpler; however it is
important to compare similar coloured LEDs or use colour correction on the measurement
results. The standard LEDs need to be calibrated by National Metrology Laboratories or a
laboratory traceable to National Metrology Laboratories.
MESURE DES DIODES ELECTROLUMINESCENTES (LED)
om
RESUME -
--
.c
`
,
,
`
,`
Ce rapport est une actualisation du rapport technique CIE 127-1997 publi

up é ant é rieurement ,
,
`,
,
`
par la CIE. Il y existe des diff é rences significatives entre les LED et les autres sources de -
`
-`
`
`
lumi è re qui ont n é cessit é , pour la CIE, l'introduction de nouvelles grandeurs avec des ,`
,
,
,
`
``
conditions de mesure pr é cis é ment d é finies pour leur caract é risation. Les nouvelles grandeurs ,
,
ro
-`
-
dé finies ici sont "l'intensit é moyenn é e d'une LED" et "le flux partiel d'une LED".
Le rapport d é crit en d é tail les conditions de mesure de l ’ intensit é moyenn é e d ’ une L
ng
"ALI" (Averaged LED Intensity), le flux total et le flux partiel d'une LED, et la distribution
spectrale de puissance. Il est montr é que les mesures par une m é thode de substitution utilisant
une LED é talon peuvent ê tre plus simples; cependant il est important de comparer des LED de
u
couleur similaire ou d'appliquer une correction de couleur aux r é sultats des mesures. Les LED
is
é talons doivent ê tre é talonn é es par un laboratoire national de m é trologie ou un laboratoire

tra?able à un laboratoire national de m é trologie.
.L
w
MESSUNG VON LEDS

w
w
ZUSAMMENFASSUNG
Dieser Technische Bericht ist eine Aktualisierung des fr ü her publizierten Technischen
Berichtes der CIE der Nummer 127-1997.
Es gibt signifikante Unterschiede zwischen LEDs und anderen Lichtquellen; diese
erfordern von der CIE die Einf ü hrung neuer Gr??en zur Charakterisierung von LEDs unter
genau festgelegten Messbedingungen. Die hier neu eingef ü hrten Gr??en sind die "mittlere
LED-Lichtst?rke" und der "LED-Teillichtstrom". Dieser Technische Bericht beschreibt im Detail
die Messbedingungen f ü r die "mittlere LED-Lichtst?rke", den "LED-Teillichtstrom", den LED-
Gesamtlichtstrom und die spektrale Strahlungsverteilung von LEDs. Es wird gezeigt, dass
Messungen nach der Substitutionsmethode unter Verwendung von LED-Normalen einfacher
sein k?nnen; hierbei ist es wichtig, LEDs ?hnlicher Farbe zu vergleichen oder aber die
Messergebnisse bez ü glich spektraler Fehlanpassung zu korrigieren. Die LED-Normale m ü ssen
bei den jeweiligen Nationalen Metrologischen Instituten oder einem Labor, das r ü ckf ü hrbar ist
auf die Nationalen Metrologischen Institute, kalibriert werden.
VI
CIE 127:2007
1. INTRODUCTION
This report is a revision of CIE 127-1997 (Measurement of LEDs) and supersedes it. CIE 127
was produced before high power LEDs became commonly available. Since CIE 127 was
published, there has been much progress in the development of the LEDs, especially of high
power LEDs in a wide range of colours including white, there have been many changes in
--`,,```,,,,````-`-`,,`,,`,`,,`--- common practice in measurement of LEDs, and also some new knowledge has become
available. This revision reflects such changes, and updates the recommendations for more
reproducible and improved measurements of LEDs.
1.1 Scope
Semiconductor devices which emit optical radiation can be divided into two distinct groups,
luminescent diodes, usually known as Light Emitting Diodes or LEDs, and laser diodes. The
present report is concerned only with the first group, LEDs. This report deals with
measurement of individual LEDs only and does not cover clusters or arrays of LEDs, fixtures
using LEDs, nor large area surface emitters such as organic light emitting diodes (OLEDs).
This report covers measurement of photometric, radiometric, and colorimetric quantities of
LEDs, to be performed in calibrating laboratories; it does not cover measurement procedures
in production lines which require other considerations. It is the responsibility of the
om
manufacturers and users to ensure that, after obtaining well characterised working standards
from their laboratory, the test set-up used for production control will measure the defined
quantities properly. The production line measurement recommendations will be dealt with in
.c
another report. The deviations from laboratory measurement conditions and possible sources
of error have to be carefully examined when the test equipment is designed and installed.
up
1.2 Terminology
ro
Strictly speaking, the term LED should only be applied to those diodes that emit visible light.
Those which emit radiation in the near infrared should, more correctly, be called IREDs
ng
(Infrared Emitting Diodes). In general, however, both groups of diodes are widely referred to
as LEDs and, since most of the measurement techniques and characterisations are common
for the two groups, the term LED is used throughout this report to cover both types. This also
u
applies to diodes emitting ultraviolet (UV) radiation. The sections relating to photometric and
is
colorimetric quantities clearly apply only to those devices emitting visible light, but if there is
any confusion this will be made clear at the appropriate point.
.L
Several terms not defined in the CIE Vocabulary are used in this document as:
w
x Averaged LED Intensity;

w
x Partial LED Flux.

w
Please see document for exact definitions.
1.3 Purpose of the report

LEDs are produced in enormous quantities and in a wide range of different types to meet the
very different specifications of a variety of applications. When a wide range of different types
of LEDs is measured, the multi-dimensional properties of the emitted optical radiation must be
considered during a measurement, not only in relation to the emitting diode but also as they
affect the receiving detector. The range of possible influences on the result of a measurement
is considerable and the related measurement uncertainty becomes correspondingly high. The
low level of the radiant power emitted by some LEDs can limit the resolution of the spectral
and spatial distribution measurements; in order to increase the signal of the detector, it has
become common practice to measure, for example, the luminous intensity of LEDs at
relatively short distances at a fairly large solid angle of the radiation coming from the LED. In
this case LEDs are not measured as a point source and measured results vary depending on
the geometrical conditions used. To minimize such variation of results, this report
standardizes such geometrical conditions so that measured values can be comparable and
reproducible among different users.
1
CIE 127:2007
Definitions of the various radiometric, photometric and colorimetric quantities used to

characterise the performance of LEDs have been collected and presented here in a way that is
intended to show some of the limiting conditions that apply during a measurement.
Recommendations are given for new CIE standard measurement conditions that can be used to
specify the properties of LEDs.
LEDs that emit visible radiation are widely used in applications where information has to
be conveyed to the human eye or for illumination purposes. This report, therefore, deals with the
characterisation of the emitted radiant power not only in terms of radiometric quantities, but also,
where applicable, in terms of photometric and colorimetric quantities. Whether radiometric or
photometric quantities are involved, they should always be measured using the appropriate SI
units.
Measurements for characterising LEDs are usually carried out using a DC current
power supply and operating under steady state conditions. The assumption is made that there is
thermal equilibrium. If the power supply is changed to multiplexed or modulated mode, even if it
is adjusted to give the LED under test the same effective electrical power consumption, the
values measured are averaged in time and the characteristics of the LEDs can be changed
significantly. The reasons for this and the possible effect on the results are discussed.
This report is based on the experience and views of the members of CIE Technical
om
Committee TC 2-34 and later its continuation, TC 2-45, but it can only represent the state of
knowledge and development in the field at the time of publication. This is a field where
production and measurement techniques are changing rapidly, and it is quite likely that future
.c
developments may render some aspects of the present report obsolete. Should it prove to be
necessary, it is hoped that the report can be revised from time to time in order to incorporate the
up
results of new developments, for example the introduction of new LEDs at shorter wavelengths,
higher light output levels, etc. to keep the document current.
ro
1.4 Categories of LED measurement

ng
LED measurements can be divided into two categories:
1.4.1 Laboratory measurements

u
Most of the manufacturers and large-scale users of LEDs first characterise the products in a
is
sophisticated laboratory. For each different type of LED, working standards are then prepared
.L
for production quality control.

w
1.4.2 Bulk testing

w
Bulk testing is used for production control or for checking the quality of incoming units. The
test set-up has to be made to operate at high speed in order to cope with large numbers of
w
units, and thus, often simplified or modified from standard measurement conditions. Such
simplified or modified measurement conditions for production control may be used as long as
the measured results are correlated to the results of laboratory measurements.
Where such routine measurements of LEDs are carried out outside a laboratory, it is of
primary importance to obtain stable, calibrated standard LEDs with the same spatial and
spectral characteristics as those of the LEDs to be tested, thus ensuring that, as far as possible,
measurements can be made on the basis of a simple, direct substitution between similar kinds
of devices.
2
--`,,```,,,,````-`-`,,`,,`,`,,`---

CIE 127:2007
2. PROPERTIES OF LEDS
2.1 Optical properties of LEDs

The radiation from a LED can be characterised by radiometric and spectroradiometric
quantities. If the LED emits visible radiation, then photometric and colorimetric quantities are
also required to quantify its effect on the human eye. Consequently, radiometric,
spectroradiometric, photometric and colorimetric quantities with their related units may all
have to be used to characterise the optical radiation emitted by a LED.
Note that for every radiometric quantity there is a photometric analogue (CIE, 1983).
The only difference is that, for the radiometric quantity, the radiation is evaluated in power units
while for photometric quantities the radiation is weighted by a spectral luminous efficiency
function, generally V( O), and multiplied by Km (= 683 lm/W). To avoid unnecessary repetition,
throughout this report, wherever the comments made apply equally to radiometric and
photometric quantities, reference is made only to the photometric quantities. If the
measurements to be made relate to a radiometric quantity, then the photometric term can be
replaced by the radiometric equivalent.
Characterisation of the optical properties of LEDs should be based upon the same
om
methods and techniques as those formulated for other types of light sources. Definitions of the
various photometric, radiometric and colorimetric quantities involved are given in the
International Lighting Vocabulary (CIE, 1987a). The basic concepts in colorimetry are described
in CIE publications (CIE, 1983; CIE, 2004). A fuller and more general treatment of the
.c
measurement of optical radiation and colour can be found, e.g., in references (Grum and
up
Bartelson, 1980; Wyszecki and Stiles, 1982).
There are hundreds of different types of LEDs available on the market, differing not only
in their spectral distribution but also in the spatial distribution of the radiation emitted, ranging
ro
from quasi-Lambertian characteristics to a nearly collimated beam with all the possible
variations in between. It is also reasonable to apply some of the quantities normally used to
ng
describe the radiation from luminaires to characterise the radiation from LEDs.
2.1.1 Spatial distribution

u
The optical radiation produced by a LED is generated by a semiconductor chip mounted in

is
some form of package. The package protects the chip during operation, incorporates the
.L
electrical contacts and supports it for handling. It should be noted that the packaging
frequently changes the spectral and spatial distribution of the radiant power emitted from the
w
chip by providing built-in reflectors or lenses and sometimes scattering material, coloured
filters or a fluorescent layer. A selection of some of the different spatial distributions of
w
luminous intensity found in LEDs is presented schematically in Fig. 1, showing the

considerable variety that can be found and the associated difficulties of defining a uniform
w
method of measurement and characterisation.
--`,,```,,,,````-`-`,,`,,`,`,,`--- 3
CIE 127:2007
30° 20 ° 10 ° 0 ° 10 ° 20 ° 30 °　 0° 10 °
Iv (- ) I v (- ) 20°
1,0
0,8 30°
0,6 40°
0,4
0,8 0,6 0,4 0,2 0 0,2 0,4 0,6 0,8 0,2 0 0,2
30° 20 ° 10 ° 0 ° 10 ° 20 ° 30 °　
0° 10 °
20°
Iv (- )
I v( - )
1,0
om
0,8 30°
0,6 40°
.c
up 50°
0,4
0,8 0,6 0,4 0,2 0 0,2 0,4 0,6 0,8 0,2 0 0,2
ro
ng
Fig. 1. Some typical spatial distributions of the luminous intensity emitted by a selection of
different LEDs. The distributions have been plotted with the maximum values normalised to unity.
u
2.1.2 Spectral distribution

is
The spectral distribution of the optical radiation emitted by LEDs is characteristic of these
.L
devices and differs in various aspects from that of other sources of optical radiation. Spectral
distribution of typical single-colour LEDs is neither monochromatic (as emitted by lasers) nor
w
broad-band (as found with incandescent lamps), but something between the two (quasi-
monochromatic), with a spectral bandwidth of some tens of nanometres. Typical relative
w
spectral distributions of LEDs for the visible region are shown in Fig. 2. Note that radiant
efficiency of LEDs varies largely depending on their peak wavelength.
w
1.2
1,2
1
1
0.8
0,8
0.6
0,6
0,4
0.4
0.2
0,2
0
0
400
400 450
450 500500 550 550 600 600 650 650 700 700
Wavelength (nm)
Wavelength (nm)
-
--
`
,
`,
,
,`
Fig. 2. Relative spectral power distributions of a series of typical LEDs. ,
`,
,
`
-`
-
``
`
`,
,
,
,`
`
`,
,
-`
-
4
CIE 127:2007
2.1.3 Area of emittance

The small packages used for LEDs offer a large variety of sizes and shapes for the light-
emitting surface. The area of emittance is characterised by its shape, size and the pattern of
the luminance across it. The luminance of the whole light-emitting surface is an averaged
value of the luminance distribution over the emitting area. Typically, the luminance is a
maximum in the centre of the exiting light beam with significantly lower values at the edges;
however there are significantly different distributions in such applications as LED lamps.
In some applications, LEDs are used under conditions where the distance between the
exit window of the package and the detector is relatively short so that the light-emitting surface
acts as an extended area and the light source can no longer be treated as a point source. In this
situation, the ratio of the illuminances produced at different distances no longer obeys the
inverse square law and the radiation pattern depends on the distance from the emitter. This is
described as the "near-field" condition. For further reading on the "near-field" condition, see
reference (Goure and Massot, 1982).
In contrast, "far-field" condition exists when the size of the emitting area is small
enough, compared to the measurement distance that the inverse square law is valid, or when
the radiation pattern is already independent of the distance from the emitter. The concepts of the
"near-field" and "far-field" conditions are discussed in Section 4.
om
2.2 Electrical characteristics
.c
2.2.1 Electrical operating conditions up
LEDs are normally operated with DC power applied in a forward bias direction and at a
constant current I F associated with a certain voltage (forward voltage) VF , which is measured
across the contacts of the LED. For accurate measurements, separate contacts for supplying
ro
current to the LED and for measuring the voltage (four-pole sockets) are recommended. They
are essential for operation at the higher currents which are typical of the single shot or
ng
multiplexed modes. The electrical power P consumed by the LED is calculated from
P = V F ・　
IF (1)
u
At low currents, the radiant power (luminous flux) rises at a rate higher than that of the
is
electrical power (start-up range). At high currents, the slope becomes flatter (saturation area),
which is mainly caused by heating of the LED chip. Under normal operating conditions (between
.L
the start up range and the saturation area), the optical radiation emitted by LEDs is linearly
correlated with the electrical current. Thus operation at constant current is recommended for
w
measurements intended to characterise the properties of LEDs.

w
--`,,```,,,,````-`-`,,`,,`,`,,`---
In many traditional light sources, a strong correlation is found between the luminous flux
emitted and the electrical power consumed. This is not so for LEDs. At constant current the
w
forward voltage of a LED decreases with increase of ambient temperatures. Adjusting the
electrical operating conditions only to stabilize the power consumed by a LED will change the
chip temperature, thus affecting the voltage drop across the LED. For this reason, stabilization
of the electrical power only is not recommended as a means for improving the stability of the
radiant output of a LED.
2.2.2 Operation of reference standards

The apparatus used to measure LED characteristics should be calibrated with LED reference
standards that have been specially selected and prepared. They should be operated at a
constant current with the temperature of the chip maintained at a constant value. When a
supplementary heating system is used to control the temperature of the chip, the LED can be
stabilised using the temperature dependent forward voltage as an indicator to be maintained
at a specified value.
The LEDs used as reference standards can be specially manufactured by incorporating
a separate resistor or transistor mounted inside the package of the LED to optimise the thermal
contact between heater and chip. Fig. 3 shows the schematics of such a reference standard.
This operating procedure is strongly recommended for all reference standard LEDs.
5
CIE 127:2007
The calibration of primary reference standards should be performed by a national

metrological institute (NMI) or a laboratory traceable to NMI. A precise description of the
measurement technique and a statement of uncertainty should be given with each calibrated
standard LED. The NMI to which the calibration is traceable should also be identified.
The reference standard LED mounted in a specially designed package should be a LED
typical of the kind to be tested. The chosen LED has to be pre-selected and "seasoned" (i.e.
burnt-in), typically for 500 hours at the ratings which are used later during operation. Accelerated
seasoning at higher currents is not recommended. It is important that the LED selected for use
as a standard shall have spectral and spatial power distributions which correspond as closely as
possible to those of the LEDs to be tested. The package should incorporate a thermostat for
keeping the reference standard LED at a predetermined temperature and provide a constant
current setting, to ensure a constant optical output.
Heat control unit Current control unit

LED lamp
Cable to power supply
om
.c
Ring for alignement
up
Fig. 3. Schematic diagram of a temperature stabilised standard LED.
ro
2.2.3 Time dependent operation

ng
In many applications, LEDs are operated under non-steady-state conditions such as

modulated current, single shot or multiplexed mode. Since the output characteristics of the
u
LED are affected by these operating conditions, it is important to identify the mode of
operation when reporting data characterising the properties of LEDs.
is
.L
2.2.3.1 Modulated current

w
An increase in current causes an increase of both the luminous output and the chip
temperature, which in turn affects the luminous output. In the case of modulated current
w
operation, the chip temperature will also fluctuate so that the average output will be different
from that obtained with steady state operation at a constant current of the same mean value.
w
Thus, the radiant efficiency he , which is the ratio of the radiant power ) e to the electrical input
power P, is a function of the average current, even if the LED is operating well within the
normal working region between start-up and saturation levels.
2.2.3.2 Pulsed operation

During production control, the measurements made to characterise the properties of each
LED are often carried out as single-shot operations within a fraction of a second and at
current levels approximating those typically used under steady state conditions. For most
LEDs the heat capacity and the heat conducting properties of the chip and package is too
large for the temperature of the chip to reach the value of steady state operation in such a
short time. Such operation thus modifies the values obtained for the LED characteristics.
Fortunately, these values from a single-shot operation are strongly correlated to the values for
steady state operation. The true characteristics can be calculated once the correlation for the
particular type of LED has been determined by a few supplementary measurements. -
-
`-
,
`,
,
,`
,
2.2.3.3 Multiplexed operation `

,,
`
`-
-
``
`
In multiplexed mode a high current is repeatedly switched on and off for a short time, the time `
,
,
,
,`
averaged value of which is equal to the normal operating DC current. As in the case of single- `
`,
,
`
-
-
6
CIE 127:2007
shot operation, the correlation has to be established between the ratio of light output to
current under multiplexed operation and the ratio of light output to current under steady state
DC operation, and this can again be established by a few supplementary measurements.
The present report is restricted to a discussion of constant current operation but the
electrical measurement methods suggested for this case can be extended to other conditions
with appropriate adjustments. The optical part of the measurement system is unchanged, but
care must be taken to ensure that the photo detector and the photocurrent measuring device
average the light linearly.
2.2.4 Forward voltage

The value of the forward voltage depends on the semiconductor material of the LED, with
variations of up to a factor of five for the different types available. At the usual working point,
with the current set to 20 mA, typical values between 1,2 V for IREDs and 6,5 V for blue LEDs
are found. The voltage V F of an individual LED also depends on the current I F and on the
junction temperature T C of the semiconductor, which can be substituted by the chip
temperature T C as a first approximation.
VF VF T C , IF (2)
om
The total derivative d V F separates the two influences.
wV F wV F
d VF dIF d T C (3)
wIF wT C
.c
up
2.2.4.1 Forward voltage dependence on current
Under stabilised temperature conditions, the relationship between the forward voltage of a
ro
LED and the current follows a well established pattern common to all semiconductor diodes.
In the normal working region, between the start-up and saturation levels, there is a close
approximation to a linear relationship with a slope given by
ng
wVF
| 10 [V/A] (4)
wIF
u
is
If the LED is operated at a working point corresponding to a current IF0 with a related forward
voltage V F0 and a differential resistance at that point given by
.L
' V F0
w
R F0 (5)
' IF0
w
then the current-voltage characteristics can be approximated by

w
§ IF ・
V F ( IF ) R F0 IF0 log ¨
¨b I 1?
? (6)
? F0 1
where
--`,,```,,,,````-`-`,,`,,`,`,,`--- § V F0 ・
b exp ¨ ? (7)
¨R F0 IF0 ?
? 1
In Fig. 4, the relationship between the forward voltage of a LED and the current is
shown at a single working point, corresponding to V F0 = 2 V and IF0 = 20 mA, for four different
values of the differential resistance.
7
CIE 127:2007
R F0 = 2 Ohm
VF (V) RF0 = 20 Ohm
I F (A)
om
Fig. 4. Relationship between forward voltage and current for a typical LED at a working point
corresponding to V F0 = 2 V and IF0 = 0,02 A shown for different values of R F0 .
.c
2.2.4.2 Forward voltage dependence on temperature up
For most LEDs, when operating at normal ambient temperatures, typical values for the
temperature coefficient of the forward voltage at a constant current are found to be in the
range
ro
wV F
| ( 1,5 to 2,5) [ m V /K ] (8)
wTC
ng
2.2.5 Ambient
--`,,```,,,,````-`-`,,`,,`,`,,`--- temperature
u
Unless otherwise specified, an ambient temperature of T amb = 25oC is assumed for LED
is
characterisation. Because of the power consumed in the LED chip, the chip temperature TC
rises after the power has been turned on and stabilises later at a value T Chip > Tamb . The rate
.L
of the temperature change depends on the level of the power input and the heat capacity and
w
thermal resistance of the LED package. After thermal equilibrium has been reached, the
value of T Chip is governed by the heat transfer to the surroundings, which takes place mainly
w
via the LED substrate (in case of older constructions via the leads of the LED). As a
consequence, the thermal properties of the electrical contacts used to supply the LED and the
w
length of the wires between chip and heat sink can significantly effect the measurement.
The temperature of the LED chip will be more or less unchanged if it is operated under
short, single-shot conditions, but a small rise in temperature is usually found during constant
current operation. Temperature effects that occur in the case of modulated or multiplexed
operation are discussed in Section 2.2.3.1 above.
2.3 Influence of temperature on the radiation
2.3.1 Shift of peak wavelength with temperature

Constant current and a temperature-stabilised voltage will result in constant consumption of
electrical power by the LED. It should be noted, however, that stabilising the power without
controlling the temperature will result in quite different operating conditions. The relative
spectral distribution of the emitted radiation will be affected in two ways. On the one hand
there will be a slight change in the shape of the distribution and on the other hand, as the
temperature rises, the whole spectral distribution might shift significantly, for GaAsP based
LEDs in the direction of longer wavelengths and for GaInN based LEDs (thus e.g. for blue
LEDs), the shift is toward shorter wavelengths. For a typical LED, this shift is about
8
CIE 127:2007
wOp
| (0,1 to 0,3) [nm/K] (9)
wTC
2.3.2 Effects of temperature on efficiency and efficacy

Small temperature changes have very little effect on the radiant efficiency of a LED. The
luminous efficacy of LEDs emitting green light is also fairly constant because the peak
wavelength of the spectral distribution is close to the maximum of the V( O) function. The
luminous efficacy of a coloured LED with a peak wavelength on the slopes of the V(O)
function is much more seriously affected by a shift in the spectral distribution. The luminous
efficacy of LEDs emitting red or blue light can, therefore, change significantly by relatively
small temperature changes. Since the spectral distribution of a LED depends on both the
power consumed and the temperature of the chip, stabilisation of current and temperature
offers the best way of controlling the operating conditions and maintaining a constant spectral
-
distribution. -
-
`
,
,`
,
`,
,
`
,,
`
-`
-
2.4 roduction tolerances `

`
`
`
,,
,
`,
Some of the most important quantities used to characterise the optical radiation from LEDs `
`
,
om
,
`
--
are related to a specific direction. It is, therefore, important to align the LEDs precisely for
these measurements. Unfortunately, there are two axes of rotation about the forward
direction; one is based on the package and the other is based on the spatial distribution of the
.c
emitted radiation. These two axes seldom coincide. The area of emittance, which can vary in
shape, size and structure, often has no well-defined limiting aperture, so that it may be
up
difficult to establish the exact location of the light centre. Taken together with typical
production
P tolerances, this results in angular and positional alignment difficulties and leads to
increased measurement uncertainty.
ro
Fig. 5 illustrates a LED for which the geometric axis of the package and the optical axis
of the emitted light do not coincide. In production testing there is not usually enough time to set
ng
the LED in the measuring jig in such a way that the luminous intensity is measured in the
direction of the optical axis. In selecting LEDs for standards it is important to use only those
u
LEDs where the optical and geometric axes coincide.

is
Mechanical
Mechanical
i
.L
w
w
w
Optical
Optical
i
Fig. 5. A LED whose mechanical and optical axes point in different directions.
3. PROPERTIES OF THE PHOTOMETER / RADIOMETER

For measurement of LEDs, a photometer or a radiometer, and/or a spectroradiometer is used
depending on whether radiometric or photometric quantities or both are to be measured.
Spectroradiometric measurements are described in Section 7. A photometer or a radiometer
for LED measurements typically consists of a detector, a filter, an entrance aperture, and an
electronic circuit to amplify and measure the detector output. For general requirements of
9
CIE 127:2007
standard photometers, refer to ref. (CIE, to be published a). Some specific information and
requirements relevant to LED measurements are described below.
3.1 Detectors
Silicon photodiodes are typically used to construct photometers and radiometers for LED
measurements. Silicon photodiodes are sensitive from ultraviolet to near infrared regions to
about 1100 nm, with a peak responsivity at about 900 nm. Silicon photodiodes typically have
linear response over several decades of input radiant flux and have nearly negligible
temperature dependence of responsivity in the visible region. Note that filters have higher
temperature dependence of their transmittance.
3.2 Angular and spatial responsivity of photometers / radiometers

A photometer and a radiometer for measurement of Averaged LED luminous/radiant intensity
do not require cosine response because light is incident from a narrow angle. They need to
have uniform responsivity only over the range of angles at which the radiation from a test LED
can be incident on the photometer or radiometer. Therefore, cosine correction is generally not
required in the front surface of a photometer or a radiometer for intensity measurement;
om
however a diffuser can be used to realize a light-sensitive area that is larger than the
detector ’ s light-sensitive area (It should be noted, that the short distance to the source
requires a significantly larger light-sensitive area of the detector than the front aperture of the
photometer). On the other hand, a photometer / radiometer used with an integrating sphere
.c
for luminous flux or radiant flux measurement requires good cosine correction.
up
For measurements of "Averaged LED Intensity" (see Section 4.3) the responsivity
across the entrance aperture of the photometer / radiometer should be uniform, to ensure that
all the radiation reaching the entrance aperture is measured with the same weight. Some LEDs
ro
have a narrow beam angle or irregular intensity distributions that can create non-uniform
illuminance distributions within the aperture. If the responsivity across the entrance aperture is
ng
not uniform, it can cause significant errors in measured Averaged LED Intensity, particularly in
CIE-B geometry, for such LEDs. A photometer with a good spatial uniformity can often be
constructed by using a non-diffuser type photometer (CIE, to published a) (it requires a large-
u
area photodiode), or by using a small integrating sphere as an input optic. Diffusers (such as
is
opal glass) are also often used for this purpose, particularly when a smaller photodiode is to be
used. It is generally more difficult to achieve good spatial uniformity by using a diffuser. A careful
.L
design and selection of diffuser materials are required to achieve sufficiently good spatial
uniformity.
w
w
3.3 Spectral responsivity of the photometers / radiometers

w
The spectral responsivity s( O) of a photometer / radiometer can be expressed by an absolute -

-
-
`
factor s 0 and a relative function s r(O) with ,

,`
,
`
,,
`
,
,
s ( O) s 0 s r ( O) (10) `
`-
-
`
``
`
,
,
,
For recommendations on the procedure for determining the spectral responsivity of optical ,`
`
`
,,
`
radiation detectors see reference (CIE, 1984a). --
If the detector is irradiated by radiation having the spectral distribution X(O), the
photocurrent i can be calculated from
f
i X 0s 0
3s ( O) S ( O) d( O)
0
r (11)
Here X(O) = X0 S(O), where X 0 is the normalisation factor and S(O) is the relative
spectral distribution. X( O) represents whichever photometric or radiometric quantity is to be
measured.
The relative spectral responsivity of a photometer should approximate as closely as
possible V( O), the CIE spectral luminous efficiency function for photopic vision (CIE, 1983). The
10
CIE 127:2007
relative spectral responsivity of a radiometer should be as flat as possible over the specified
spectral range.
3.3.1 Photometer to measure white LEDs

Commercially available photometers are usually classified according to their f1' number (CIE,
1987b). The f1 ' is recommended for measurement of white LEDs. Unless the spectral
mismatch correction is always applied (as described in Section 5.2), it is recommended that a
photometer used to measure white LEDs have an f1' value < 3,0 %. The f1 ' is defined as
f '1 3s * ( O) V ( O) dO (12)
rel
3V ( O)d O
where s*( O) rel is the normalized relative spectral responsivity of the detector:
s * (O)rel s ( O)rel 3S ( O) A V ( O)dO (13)

3S( O) A s (O)rel dO
S(O)A is the relative spectral distribution of CIE standard Illuminant A. The latter is included to
om
take into account the fact that photometers are normally calibrated using a tungsten filament
lamp set to the distribution temperature of CIE standard Illuminant A. Errors for white LEDs
will be minimized if f1' is small, but uncertainty still needs to be properly evaluated.
.c
If photometers used for LED measurements do not meet these recommendations of f 1',
the use of such photometers should be limited to strict substitution (comparison of the same
up
type of standard and test LEDs having the same colour), or such photometers be furnished with
individually measured relative spectral responsivity data so that a spectral mismatch correction
(see Section 5.2) can be applied.
ro
3.3.2 Photometer to measure coloured (non-white) LEDs

ng
In the case of single-colour LEDs, the spectral mismatch errors can be very large even if f1 ' is
reasonably small, due to the fact that some LED spectra are peaking at the wings of the V( O)
u
function where the deviation makes little effects on f1 ' but can cause large errors.
is
Determination of a better and more useful number for the goodness of the fit to the V( O)
.L
function for LEDs is beyond the scope of this report. [From preliminary results it seems that one
number similar to f1 ' is not sufficient to evaluate the accuracy of photometers for measurements
w
of all of different colour types of LEDs; instead maybe as many as 4 numbers are needed (Csuti
et al., to be published).]
w
For photometers to measure single colour LEDs, it is recommended that the relative
w
spectral responsivity of photometer be supplied, with examples shown how to apply correction
for spectral mismatch errors and how to evaluate the measurement uncertainties of the
measured photometric quantity of a given coloured LED.
4. QUANTITIES DEFINING SPATIAL RELATIONS
4.1 Normalisation factor and relative spatial distribution
In general, the luminous intensity I(T, I ) depends on the direction ( T, I ) and this dependence
is called the spatial intensity distribution. It should be noted that measurements of luminous
intensity, including those required to map the spatial distribution, must be made over a very
small element of solid angle d : and this requires a detector where the diameter of the input
aperture and the diameter of the source are small compared to the distance from the source.
If the absolute value of the intensity I( T, I ) is measured in a specified reference direction
corresponding to T = T0 and I = I 0 and denoted by I00 = I( T0 , I 0 ), then this can be used as a
normalising factor and a relative spatial intensity distribution G(T, I ) defined. The spatial
intensity distribution I( T, I ) can be expressed as
11
--`,,```,,,,````-`-`,,`,,`,`,,`---

CIE 127:2007
I (T, I ) = I 00 . G( T, I ) (14)
which can be rewritten as
I (T , I )
G (T , I ) (15)
I 00
For a spatial intensity distribution there is no dependence on angle I at angles T = 0 and

T = S. Consequently, the value in the direction T = 0 is the one usually preferred for
normalisation, making I00 = I( T = 0).
The simplest form of the function G(T, I ) is
G(T) = G (16)
where G is a constant. This represents the spherical spatial distribution of a totally isotropic
point source.
Another spatial distribution that is easily expressed mathematically is the Lambertian
distribution. With T measured as the angle between the direction considered and the
perpendicular to the surface, the spatial distribution for all values of I is given by
om
G (T ) G0 cos T (17)
where the range of angles is limited to a hemisphere with 0 d T d S / 2. This spatial distribution
.c
is normally used as a reference.
It is not possible to express most practical spatial distributions in terms of a simple
up
mathematical function, but symmetrical spatial distributions are often characterized by specifying
the angles corresponding to 50 % and 10 % of the maximum value (CIE, 1987b). It should be
ro
noted that, for structured intensity distribution curves, there may be more than one angle that
produces the 50 % or the 10 % value. When this method is used, it is recommended that the
first angles for such intensity values starting from 0 ° (measured from direction of mechanical
ng
axis) be reported.
The majority of LEDs are designed to provide a distribution with the maximum intensity
u
in the direction T = 0, but this is not always the case and for some LEDs the construction of the
is
device gives a significantly lower value in the direction of the geometrical axis than for some off-
axis angles. One of the examples in Fig. 1 shows this effect.
.L
Sometimes, because of production tolerances, even if the LED is mounted in a

w
cylindrical package, the mechanical axis of the package (which is used to align the LED in the
measurement apparatus) and the optical axis (which is the axis of rotational symmetry of the
w
spatial distribution) may have slightly different directions (see Fig. 5). The measurement
procedure must take account of the influence that this could have on the results.
w
By no means all production LEDs have a spatial distribution that shows perfect axial
symmetry. Fig. 6 shows two common forms of asymmetric spatial distribution that are
sometimes found in LEDs and can lead to alignment problems. The spatial distribution of the
LED depicted in Fig. 6a shows a small minimum in the direction of the package axis ( T = 0 ° ),
and a maximum in an off-axis direction. Fig. 6b shows I( I ) plotted at constant T f or a LED in
which the non-circular shape of the intensity distribution indicates the departure from rotational
symmetry.
12
--`,,```,,,,````-`-`,,`,,`,`,,`---

CIE 127:2007
I T = 0 ° 10 ° 20 °
1,0 30° I
0,8 40°
50°
0,6
60°
0,4 0,2 0 0,2 0,4 I
(a) (b)
Fig. 6. Two frequently occurring non-symmetrical intensity distributions: (a) the optical axis is
off the geometric one, (b) the spatial intensity distribution is non axial symmetric.
om
4.2 Measurement of directional quantities
4.2.1 Luminous intensity
.c
Luminous intensity is defined as the quotient of the luminous flux d I v leaving the source and
propagated in the element of solid angle d : containing the given direction, by the element of
up
solid angle.
I dI v / d: (18)
ro
Although this may appear, at first sight, to be simply a question of making a

measurement of the luminous flux per solid angle in a given direction, in reality the situation is
ng
often far more complex. The concept of luminous intensity requires the assumption of a point
source, or at least a source small enough for its dimensions to be negligible compared to the
u
distance between source and detector and, in principle at least, there is also a requirement that
the measurement should be made over a very small element of solid angle.
is
Many LEDs have a relatively extended area of emittance (see Section 2.1.3) which, at
.L
the short distances at which they are often measured, may be too large to be treated as a point
source. In addition, the package of LEDs often has a lens and shifts the effective centre of
w
emission.
w
4.2.2 Illuminance
w
The illuminance E V (T, I ), produced at a distance d from a source in a direction ( T, I ) on an

element of surface normal to that direction, is related to the luminous intensity I V ( T, I ), in that
direction by the equation
2
E V (T, I ) = I V ( T, I ) / d (19)
provided again that the distance is large enough for the source to behave effectively as a
point source and that the angle subtended by the detector is at least small enough for the
illuminance to be effectively uniform. Eq. 19 is known as the "inverse square law", but it can
be rewritten as
2
IV (T, I ) = E V ( T, I ) ・　
d (20)
This is the basis of all practical measurements of luminous intensity. The quantity
actually measured is the illuminance at the surface of a photometer and the intensity is then
calculated on the basis of Eq. 20 by multiplying the illuminance by the square of the distance
from the source.
For accurate measurements of luminous intensity, however, not only must the relative
size of the source and the angle subtended by the detector be small, but it is also important to
be able to measure the exact distance between the source and the photometer. Since the actual
13
--`,,```,,,,````-`-`,,`,,`,`,,`---

CIE 127:2007
location of the effective light centre of a LED can be difficult to determine due to its lens or total
diffusivity, distances are often measured from an arbitrary location on the LED package.
4.2.3 Location of the effective emitting surface

If the measurement distance is large enough, the actual position of the reference point
shouldn't matter very much, but because of the large variety of different types of LEDs
available, no simple general rule can be laid down to determine the minimum safe distance
for accurate measurement (see CIE, 1987c). This is the reason that CIE recommends the use
of the concept of "Averaged LED Intensity" (Section 4.3).
4.2.4 "Near-field" and "far-field" measurement conditions

If a true luminous intensity is to be measured, the size of the emitting area of the source and
of the receiving surface of the photometer must be small enough to be insignificant compared
to the distance between the two. In this situation, the inverse square law will be obeyed and
2
the illuminance E V at the surface of the detector will be given by EV = IV / d (Eq. 19) where IV
is the luminous intensity of the source in the given direction and d the distance between the
light centre of the source and the detector. This is sometimes referred to as the "far-field"
condition.
om
In many applications, however, measurements are made on LEDs at relatively short
distances, where either the relative size of the source is too great for it to be treated as a point
source or the angle subtended by the detector at the source becomes too large. This is known
.c
as the "near-field" condition. The inverse square law can no longer be applied and the
illuminance measured by the detector becomes critically dependent on the exact measurement
up
conditions.
ro
4.3 Averaged LED Intensity

In manufacturers ’ literature, one of the parameters most commonly quoted as a measure of
ng
the directional output of a LED is luminous intensity. Unfortunately, in many cases, the term is
incorrectly used and the quantity measured is not really a true intensity as defined in Section
4.2.1.
u
The actual procedure employed is to make a measurement of the flux incident on a

is
detector at a measured distance from the LED and to calculate the solid angle by dividing the
.L
area of the detector by the distance squared. Because these measurements are commonly
made at relatively short distances, the emitting area of the LED could, in many cases, be large
w
enough compared to the distance from the detector to act as an extended area rather than as a
point source. This is the situation known as the "near-field condition" as described above. It is
w
also possible, if the detector is too close to the source, that the value of the true luminous
intensity may vary as viewed from different parts of the detector surface.
w
In situations of this kind, which are very common in the real world of LED measurement,
the quantity measured is not intensity in the traditional sense but represents a form of Averaged
LED Intensity; averaged that is for the various individual elements that make up the extended
area of the emitting surface of the LED as well as over the different parts of the detector surface.
Unfortunately, this distinction is not just a quibble over the exact wording of a definition. There is
a real problem because, in this situation, the results of the measurements and the applicability of
the measured values are critically dependent on the exact conditions under which the
measurement has been made. This makes it very important to agree and define a precise
measurement geometry that can be applied to a wide range of LEDs in order to allow a true
comparison between different products and, equally important, between similar products from
different manufacturers.
In an attempt to offer a solution to this problem, the CIE has decided to recommend the
adoption of a new term, specific to LED measurements, to describe the quantity measured
under such "near-field" conditions and to define two standard measurement geometries
associated with it. The two measurement geometries are based on current practice in the
industry and on views expressed by both manufacturers and users of LEDs.
14 --`,,```,,,,````-`-`,,`,,`,`,,`---

CIE 127:2007
The new term is called the Averaged LED Intensity. (Averaged LED luminous intensity
or Averaged LED radiant intensity).
The measurement geometries are designated as CIE Standard Conditions A and B for
the measurement of LEDs. For Averaged LED Intensities determined under these conditions the
symbols ILED A and I LED B are recommended. They can be used for either radiometric or
photometric quantities (e.g. I LED A e , ILED B v ).
Both conditions involve the use of a detector with a circular entrance aperture having an
2
area of 100 mm (corresponding to a diameter of about 11,3 mm). The LED should be
positioned facing the detector and aligned so that the mechanical axis of the LED passes
through the centre of the detector aperture. It is the distance between LED and detector that
constitutes the difference between conditions A and B. The distances are:
for CIE Standard Condition A: 316 mm, and
for CIE Standard Condition B: 100 mm.
In both cases the distance is measured from the front tip of the LED to the plane of the
entrance aperture of the photometer or radiometer.
If the detector has been calibrated for illuminance, the Averaged LED luminous intensity
om
can then be calculated from the relation
2
ILED v = EV ・　
d (21)
where EV is the average illuminance in lx measured by the detector and d the distance,
.c
expressed in metres. For Condition A, d = 0,316 m and for Condition B,
up d = 0,100 m.
These conditions correspond to solid angles of view of 0,001 sr for Condition A and
0,01 sr for Condition B, but the actual dimensions are as important as the angles in ensuring
consistent results. The equivalent full plane angles are approximately 2 ° for Condition A and
ro
6,5 ° for Condition B.

-
-
ng
`-
,
,
,`
`
,,
`
,,
`
-
`
-
`
``
LED
u
`
,,
Detector ,
,
`
`
`
is
,,
`
--
.L
w
Circular aperture
2
of area, A = 100 mm
w
Filter
d
w
Fig. 7. Schematic diagram of CIE Standard Conditions for the measurement of Averaged
LED Intensity. Distance d = 316 mm for Condition A, d = 100 mm for Condition B.
4.4. Measurement of spatial and directional properties

It is desirable that the LEDs selected for use as working standards should have a relative
spatial distribution of intensity similar to that of the test LEDs to be measured. For luminous
flux measurements using an integrating sphere, use of standard LEDs having similar beam
characteristics will minimize spatial non-uniformity errors of the integrating sphere. For
intensity measurements, this requirement is not so critical, but the use of standard LEDs
having similar beam characteristics will help reduce stray light errors (errors due to light
entering the photometer not directly from the source). Standard LEDs for intensity
measurements should be selected such that the optical axis of the beam is approximately on
the mechanical axis and the spatial distribution around the center axis (within ± ~10 ° ) is
15
CIE 127:2007
smooth and fairly constant so that alignment error will not cause large measurement
uncertainties. Thus LEDs having a very narrow beam or having some structures in intensity
distribution curves should be avoided. A measurement to check these spatial intensity
distribution characteristics should be carried out as a first step in selecting suitable standard
LEDs. If the laboratory is equipped with a goniophotometer, it should be used to perform a
direct measurement of the spatial distribution of intensity. The best arrangement is to position
the front tip of the LED at the centre of the goniophotometer and measure the radiation from
as large distance as the instrument will allow.
-
In laboratories where a goniophotometer is not available, a simple test would be to -

-
,`
,
,`
illuminate a white sheet of paper with the LEDs and compare the beam pattern visually. `
,
,
`
,
`,
-
`-
`
`
`
`
,
,,
,
``
5. MEASUREMENT OF AVERAGED LED INTENSITY `

,,
`
-
-
5.1 Substitution method

A test LED is calibrated in comparison with a reference LED of the same type (similar spectral
distribution). The reference LED must have been calibrated for the same geometry (CIE
Condition A or Condition B) as the test LED is to be measured.
om
The Averaged LED Intensity ILED [cd] of the test LED is obtained by
y test
I LED, test I LED, ref (22)
.c
y ref up
where ILED, ref and ILED, test are the Averaged LED Intensity of the reference LED and the test
LED, respectively. yref and ytest are the photometer signals for the reference LED and for the
test LED, respectively.
ro
With such a strict substitution method, there is no need for spectral mismatch correction
ng
and the measurement is most simple. However, if many different types of test LEDs are
measured, many different types of standard LEDs are needed. (Also, there will still be some
spectral mismatch errors due to small difference in spectral distribution between the reference
u
LED and test LED, which should be evaluated as an uncertainty component.)

is
5.1.1 Substitution with fewer standards

.L
In many cases, there are too many different types (colours) of LEDs to be measured and so
w
many standard LEDs cannot be maintained. In such cases, the method described in Section
5.2 is acceptable. In this method, it is recommended that calibrated standard LEDs (traceable
w
to a national standardizing laboratory) of several colours of interest are measured using the
user ’ s set up and the results are compared to verify the uncertainties of measurements.
w
5.2 Applying spectral mismatch correction

This method requires the knowledge of relative spectral responsivity of the photometer head.
The photometer head is calibrated with standard LEDs of a certain colour (e.g., green or
white), and test LEDs of any other colours are measured with spectral mismatch correction
applied as below.
y test
I LED, test F ILED, ref (23)
y ref
Here F is the spectral mismatch correction factor calculated for each test LED as given by
F 3S ( O)V ( O) dO 3S (O) s
t r rel ( O) d O
(24)
S ( O)V ( O) d O S (O) s rel (O) d O
3 r
3 t
where
16
CIE 127:2007
St (O) is the relative spectral distribution of the test LED;

Sr(O) is the relative spectral distribution of the reference LED;
s rel ( O) is the relative spectral responsivity of the photometer head, and

V(O) is the CIE spectral luminous efficiency function of the photopic vision.
The correction factor F is a multiplier to the measured signal of the photometer head for
the test LED.
5.3 Use of a spectroradiometer

A spectroradiometer may be used in place of the photometer head for Averaged LED
Intensity measurements if the spectroradiometer is appropriately designed for the LED
measurements. See Section 7.4.1.1.
5.4 Detector-referenced method

Rather than using standard LEDs to calibrate the photometer head, one can use a
photometer as a reference standard (called a reference photometer), as is common practice
om
in general photometry. Selected high-quality photometers are stable over a long period of
time. A photometer head having the required aperture for Averaged LED Intensity is
calibrated for illuminance responsivity [A/lx] for a reference source spectrum (typically,
.c
CIE Standard Illuminant A) at the distances corresponding to CIE Condition A and CIE
Condition B geometry. The responsivities for CIE Condition A and B can be slightly different
up
due to near-field effects. Such calibration of a photometer head may be available from a NMI,
and also, it can be performed by users by calibrating the photometer head against a standard
LED traceable to a NMI. When the photometer head is placed at the exact distance d (316
ro
mm or 100 mm for CIE Condition A or B geometry), the photometer head can measure the
Averaged LED Intensity of the test LED directly by
ng
2 y
I LED A F d ; d = 0,316 [m] (25)
s LED A
u
or
is
y
.L
2
I LED B F d ; d = 0,100 [m] (26)
s LED B
w
where y is the signal of the photometer head, s LED A and s LED B are illuminance responsivities
w
of the photometer head for CIE Condition A and Condition B, respectively, and F is the
spectral mismatch correction factor.
w
The spectral mismatch correction factor F is calculated according to Eq. 24. The
difference of this method from the method described in Section 5.2 is that, in this detector-
referenced method the scale is maintained on the photometer head. The responsivity is
maintained for only one reference source, and LEDs are always measured with spectral
mismatch correction applied. The relative spectral responsivity of the photometer head must be
known for this method, as well as the relative spectral power distributions of the test LEDs. The
photometer head should be recalibrated periodically to maintain a required uncertainty.
When this method is applied, it is recommended that calibrated standard LEDs
(traceable to a NMI) of several colours of interest are measured using the user ’ s set up and
compare the results to verify the uncertainties of measurements for LEDs of various colours.
--`,,```,,,,````-`-`,,`,,`,`,,`---
6. MEASUREMENT OF LUMINOUS FLUX
6.1 Measured quantities

The term, total flux, means all the flux emitted from the source, therefore, the total flux
integrated over all the directions (4 S steradian solid angle). However, this is not always the
17
CIE 127:2007
important quantity that the application of a particular LED requires. In some cases, the
concept of partial flux radiated into a certain solid angle is needed, ignoring the flux in
unintended directions (for example, backwards). Therefore, in addition to total luminous flux,
a new quantity, "Partial LED Flux" is introduced for LED measurements.
6.1.1 Total luminous flux

Total luminous flux is the fundamental quantity for a light source. It is defined as the
cumulative luminous flux of a light source for the solid angle 4 S steradian. The symbol of total
luminous flux is ) or ) v and the unit is lumen. It is defined as the integral of luminous intensity
over the entire full solid angle from the source,
?
3 I d:
:
(27)
or, the integral of illuminance from the source over the entire area of a closed imaginary
surface surrounding the light source,
?
3E d A
A
(28)
The total luminous flux of a LED, therefore, should include all the flux emitted from a
om
--`,,```,,,,````-`-`,,`,,`,`,,`---
LED including backward flux.
6.1.2 Partial LED Flux
.c
Partial LED Flux is a quantity used for specific applications
of LEDs. It is defined as the flux leaving the LED and
up
propagating within a given cone angle (centered from the
LED's mechanical axis) that is determined by a circular )) LED, x
LED, x
ro
aperture of 50 mm diameter and the distance measured

from the tip of LED. Fig. 8 illustrates this definition. 50 mm diam.
ng
Distance d is set for a desired cone angle x° as given by
25
d >mm @(29) d
u
x
tan x°
x
is
2
.L
where 0 °
d x d 180 °　
The symbol for this quantity is ) LED, x , with the value
w
of x being the cone angle (diameter) in degrees. For

w
example, ) LED, 180 , corresponds to the flux emitted in the

front half hemisphere (forward flux) in which case d = 0. Any
w
flux emitted in the directions other than in the given cone

angle is ignored.
The reference point of the LED is the tip of the
enclosure of the LED, though it is not the effective center of
light emission, because it can be easily identified for any
type of LED while the effective center of emission is difficult
to determine and sometimes unknown. It is chosen for
simplicity and reproducibility of measurement. The diameter Fig. 8. Partial LED Flux.
of the aperture (50 mm) is fixed in order to achieve reproducibility in measurement.
(Measurement results would vary if apertures of different sizes were used for the same cone
angle.) This is not a real partial flux defined in the far field; rather, it is called "Partial LED Flux", a
quantity that can be used for practical measurement of LEDs with simple instrumentation and
reproducible results.
It is recommended that total luminous flux (Section 6.1.1) is used as much as possible.
Partial LED Flux is used only when total luminous flux does not satisfy the need for a given
application. When Partial LED Flux is used, it is recommended that such full angles as 40 ° , 60 °
90° and120 ° are used as much as possible. Use of many different angles would make data
comparison difficult. Also, it is preferable that the angle be chosen such that most of the main
18
CIE 127:2007
beam flux from the LED is contained in that solid angle. Such a condition will make the
measurement less sensitive to LED alignment errors and the errors in distance and aperture
diameter.
In a production environment, Partial LED Flux may be measured in a modified geometry
(smaller diameter of the aperture), in which case, the measured results can be equalized or
corrected to the values under the defined condition (50 mm aperture) based on correlation of
results between the two geometries for given type of LED, with an additional uncertainty taken
into account.
6.2 Methods of flux measurement

For measurement of total luminous flux, goniophotometers or integrating spheres are used.
For measurement of Partial LED Flux, integrating spheres are commonly used.
6.2.1 Goniophotometer method
6.2.1.1 Total luminous flux measurement

A goniophotometer is a device to measure the luminous intensity of the source (or illuminance
om
from the source at a given distance) in many different directions from the source. Compared
with the sphere method, the goniophotometer method is theoretically free from errors due to
differences in intensity distribution of the light source under test. It does not require total
luminous flux standards. Instead, it requires longer time for measurement of each sample.
.c
By measuring the luminous intensity distribution I (T, I ) of the source, the total luminous
up
flux is obtained by
2S S -
ro
-
) I (T , I ) sin T d T d ) (30) -`
3I 3T 0 0
,
`,
,
`
,
,
`
,,
`
-`
ng
A goniophotometer can be configured such that the illuminance distribution E ( T,I ) is -

`
`
`
`
,,
measured over an imaginary spherical surface with radius r [m], rather than measuring luminous ,
,`
`
`
,,
intensity. In this case, the total luminous flux is given by `

--
u
2S S
2
is
) r E (T, I ) sin T d T d I (31)

3I 3T
0 0
.L
By the definition given in Eq. 31 the location of the light source in the imaginary sphere
does not matter, Therefore, theoretically, the alignment of light source (LED) is not relevant to
w
the measured total luminous flux, though the light source is normally placed at the center of
w
rotation of the photometer.

The angle intervals of measurement should be carefully chosen depending on the
w
sharpness of the beam pattern of the LED. In reality LEDs are not point sources and they have
non-uniform light distribution. The measurement distance (radius of rotation of photometer)
should be set long enough (typically 300 mm or longer) so that errors in distance measurement
(including error in position of photometer reference plane) (when Eq. 31 is used) or errors in
LED alignment (when Eq. 30 is used) will be negligible. If a goniophotometer is also designed to
measure Averaged LED Intensity, distance r can be chosen as those for Conditions A and B
(100 mm or 316 mm). It should be noted that the angular resolution under these conditions is
fairly low (due to the relatively large solid angle in which flux is measured).
The photometer head of the goniophotometer should meet the spectral responsivity
requirement given in Section 3.3. The calibration of the photometer head (and spectral
mismatch correction) should follow the recommendation for Averaged LED Intensity
measurement (Section 5). The instrument also requires careful shielding of ambient light and
reflected light within the instrument.
The range of angular scan must cover the entire solid angle to which the test LED emits
light. Note that some LEDs have significant amount of backward emission even though they
have a narrow beam pattern in forward direction. Such backward emission must be included for
total luminous flux. Note that some goniophotometers can scan only the front hemisphere, in
which case, any backward emission is ignored, leading to some error in total luminous flux.
19
CIE 127:2007
6.2.2 Integrating sphere method
6.2.2.1 Total luminous flux measurement

A simpler way to measure the total luminous flux of a LED is to use an integrating sphere
photometer. It is a device to perform spatial integration of flux optically, thus the total luminous
flux can be measured with one fixed photometer head and measurement can be instant. An
integrating sphere photometer is calibrated with a total luminous flux standard. A test light
source is measured by comparison to a standard source having similar spatial and spectral
distributions. Therefore, this method requires standard LEDs calibrated for total luminous flux.
Compared with goniophotometer, measurement is fast, but it is liable to errors when the
spatial intensity distributions of test LED and standard LED are dissimilar. This type of error is
difficult to correct, so the error should be minimized by using well-designed sphere geometry
and similar type of standard LEDs as the test LED.
Fig. 9 shows recommended sphere geometries for total luminous flux measurement of
LEDs. Geometry (a) is recommended for all types of LEDs including those having a narrow
beam profile or those having broad and backward emissions. This sphere geometry provides
good spatial uniformity of responsivity over the sphere wall and is less sensitive to the
differences in spatial intensity distribution of a LED. Under the old practice, a test LED was
om
mounted on the sphere wall, which is generally not recommended for total flux measurement,
due to the loss of backward emission. However geometry (b) is acceptable for LEDs having no
backward emission. This geometry has an advantage that the test LED can be easily mounted
on the sphere wall. Note that 5 mm epoxy type LEDs can have significant amount of backward
.c
emission and, therefore, should use geometry (a). High-power LEDs having a large heat sink
and no backward emission can be measured with geometry (b) where only the LED head is
up
inserted into the sphere and the large heat sink stays outside the sphere.
Cosine-corrected Cosine-corrected
ro
photometer head photometer head

ng
baffle
baffle
u
is
Test LED Test LED

.L
Substitution
Substitution
w
w
Standard LED
w
Standard LED
Auxiliary LED
Auxiliary LED
(a) (b)
Fig. 9. Recommended sphere geometries for LED total luminous flux.

(a) for all type of LEDs, (b) for LEDs with no backward emission
For either geometry, a minimum diameter of 20 cm is recommended for the sphere.

Spheres of 20 cm to 50 cm diameter are commonly used for LED measurement. The larger the
sphere, the less are errors for spatial nonuniformity (due to relatively smaller size of the baffle)
and the sensitivity to self-absorption is less, but the signal is also less. High-power LEDs (e.g.,
> 1 W consumed electrical power) can be measured in a large integrating sphere (e.g., a 2 m
sphere used for measurement of traditional lamps), in which the whole unit of the LED and heat
sink can be mounted in the center of the sphere in geometry (a).
The sphere must be equipped with an auxiliary LED (CIE, 1989). Self-absorption
measurement must be done unless the test and standard LEDs are of the same type and size
(strict substitution), or unless the sphere size is so large that the self-absorption by a LED is
20
--`,,```,,,,````-`-`,,`,,`,`,,`---

CIE 127:2007
negligible. Self-absorption may be different for different colours of LEDs, depending on the
sphere characteristics. It is best to use an auxiliary LED having similar colour as the test LED if
the differences are found to be significant.
An interior coating reflectance of 90% to 98% is preferred, depending on the sphere
size and usage of the sphere. The higher the reflectance, the higher the signal obtained, and the
less the errors associated with LED intensity distribution variations. However, with higher
reflectance, the sphere responsivity is more sensitive to self-absorption effects, long-term drift
and spectral power distribution differences. The size of the baffle should be as small as possible
to shield the photometer from direct illumination from reference and test LEDs of the largest size
measured.
In either geometry (a) or (b), it is important that the photometer head has a good cosine
response and good V(O) match. The spectral responsivity requirements (Section 3.3) apply to
the total sphere system (photometer head + the integrating sphere). See Section 6.2.3.3 for
spectral mismatch correction methods. If spectral mismatch errors are not corrected, strict
substitution (Section 6.2.3.1) should be performed, where test LEDs are compared with
standard LEDs of the same type (nearly the same spectral distribution). Care must be taken, if
LEDs are measured, which are able to generate fluorescence in the sphere. The error can be
significant when the LED peak wavelength is in the wings of V(O) (e.g., deep blue LEDs) and
om
fluorescence occurs in the green region where the value of V(O) is high, in which case, the effect
of fluorescence is magnified in the measured luminous flux.
6.2.2.2 Partial LED Flux measurement
.c
An integrating sphere photometer with a geometry depicted in Fig. 10 is recommended for
up
Partial LED Flux measurement. The sphere has an opening, to which a precision aperture
(50 mm diameter) is attached. The reference plane (the plane containing the knife edge) -
-
-
,`
should be placed to be flush with the inner surface of the sphere (to allow full cone angle up
ro
,
,`
`
,,
to x = 180 ° ). The area of the aperture must be measured with a stated uncertainty, since it `
,
,
`
-
`-
directly affects the uncertainty of Partial LED Flux measurement. It is recommended that the `
``
ng
`
,
,
size of the sphere be 20 cm or larger in diameter. A relatively high reflectance coating (95% ,
,
``
`
,,
to 98%) is preferred for higher signal and better spatial uniformity of sphere responsivity, `
-
-
considering a loss of effective reflectance of the sphere due to the large opening. The
u
distance d (from the tip of the LED to the reference plane of the aperture) is determined for a
is
given full cone angle x° , according to Eq. 29.

.L
The baffle is placed about half way between the photometer head and the opening, and
it should be smallest possible but large enough to shield the photometer (light sensitive area)
w
from the entire opening of the aperture.

w
Self-absorption measurement will not be needed if the test LED is small and is placed
far from the opening (e.g., x = 60° ). However, self-absorption measurement may be needed if
w
the test LED (including its mount) is large and/or it is placed close to or at the opening (e.g.,
d = 0, x = 180 ° ).
The area around the test LED and the opening must be shielded from ambient light, and
care should be taken so that only direct light from the test LED can enter the opening, without
any stray light or reflections from other objects around the LED.
21
CIE 127:2007
Cosine-corrected
photometer head
dd
baffle
Test LED
50 mm
Substitution
Precision
--`,,```,,,,````-`-`,,`,,`,`,,`---
Standard LED
aperture
Auxiliary LED
om
Fig. 10. Recommended sphere geometry for Partial LED Flux measurement.
The test LED is measured in substitution with a standard LED of the same type. Use of
.c
standard LEDs having similar spatial intensity distribution is recommended, as it will cancel out
or reduce many sources of errors such as stray light, distance d, and aperture area A. See
up
Section 6.2.3 for the use of standard LEDs and spectral mismatch corrections.
6.2.3 Methods for sphere calibration and correction

ro
6.2.3.1 Strict substitution

ng
When a test LED of a certain colour (spectral distribution) is to be measured, the integrating
sphere photometer is calibrated against a standard LED of the same colour (or nearly the
u
same spectral distribution). When measuring a test LED of another colour, the sphere
photometer is calibrated against another standard LED of the same colour. With such a strict
is
substitution method, there is no need for spectral mismatch correction, and the measurement
.L
is most simple. However, if many different types (colours) of test LED are measured, many
different standard LEDs are needed. (Also, note that, even in such strict substitution, the
w
spectral mismatch errors will not be zero due to small differences in spectral distribution
between the reference and test LEDs, and such errors should be evaluated as an uncertainty
w
component). Auxiliary lamps should be used to determine the self absorption for those cases
where the test LEDs differ from the standard LED, as the colour and spatial distribution is
w
immaterial: during operating the auxiliary LED the LED under test is switched off. Important is
that the two are different, i.e. their absorption differs.
Substitution with fewer standards

In many cases, there are too many different types (colours) of LEDs to be measured and so
many standard LEDs cannot be maintained. In such cases, the method described in Section
6.2.3.2 is recommended. In this case, it is recommended that calibrated standard LEDs
(traceable to a NMI) of several colours are measured using the user ’ s set up and to compare
the results to verify the agreement of the results within the stated uncertainties for each colour
of LED.
6.2.3.2 Applying spectral mismatch correction

If the relative spectral responsivity of the integrating sphere photometer is known, the sphere
photometer is calibrated with standard LEDs of a certain colour (green, white, etc.), then the
test LEDs of whatever colour are measured with spectral mismatch correction applied. The
relative spectral distribution of the test LED must be measured if not known. The spectral
mismatch correction factor F is calculated by Eq. 24, with s rel ( O) replaced by
22
CIE 127:2007
s rel (O) s ph, rel (O) Trel ( O) (32)
where s ph, rel ( O) is the relative spectral responsivity of the photometer head, and Trel ( O) the
relative spectral throughput of the integrating sphere.
P
Measurement of relative spectral throughput
The relative spectral throughput T rel (O) of an integrating sphere can be measured using the
following methods:
1) For total flux measurement in integrating spheres, first, measure the spectral
distribution of a reference tungsten lamp outside the sphere, with a
spectroradiometer. Measure in several different directions to check that the lamp has
spatially uniform spectral distribution. The spectroradiometer must be in irradiance
mode and must have good cosine-correction. Then, operate the lamp in the sphere
and measure the spectral distribution at a detector port (with direct light shielded)
using the same spectroradiometer. The ratio of the measured spectral distribution
inside the sphere to the data outside the sphere will give the relative spectral
throughput of the sphere.
2) Using integrating spheres for Partial LED Flux measurement, introduce a beam of
om
light from a reference tungsten lamp. Measure the spectral distribution of the lamp
with a spectroradiometer (irradiance mode) in the direction of the entrance opening.
Then, measure the spectral distribution at the detector port of the integrating sphere
.c
using the same spectroradiometer. The spectroradiometer must have good cosine-
correction. The ratio of the measured spectral distribution at the sphere detector port
up
to the data measured directly of the lamp will give the relative spectral throughput of
the sphere.
ro
Note that this method is susceptible to errors if the sphere coating is inhomogeneous
and/or has fluorescence. The tungsten lamp to be used in this measurement should be small
ng
enough relative to the size of the sphere (e.g., a miniature tungsten halogen lamp may be
appropriate for a 20 cm sphere).
It is possible to determine the sphere throughput T rel (O) via calculation also:
u
is
U(O)
T rel (O) k (33)
1 U(O)
.L
where U( O) is the spectral reflectance of the sphere coating, and k is a normalizing constant.
w
Note that the coating ages and its surface is being contaminated as the sphere is used. Thus,
the results calculated from data of a coating sample may not be accurate for actual
w
integrating spheres. The U( O) must be measured in hemispherical illumination geometry. In

w
addition, due to the denominator, 1- U(O), measurement errors in U(O) tend to be magnified.
6.2.3.3 Use of a spectroradiometer

A spectroradiometer may be used as the detector for the integrating sphere, and the total
luminous flux can be measured without spectral mismatch correction if the spectroradiometer
is configured to measure total spectral radiant flux of LEDs. (See Section 7.4.2.1.)
--`,,```,,,,````-`-`,,`,,`,`,,`---
7. S ECTRAL MEASUREMENT
7.1 The concept of spectral distribution
7.1.1 Spectral concentration

For any given radiometric quantity Xe , the spectral concentration of that quantity is the
differential of the quantity with respect to wavelength O and is given by
d X e ( O)
X O( O) (34)
dO
23
CIE 127:2007
XO(O) is also known as the spectral distribution of that quantity. This function is a
wavelength dependent function. The unit of the spectroradiometric quantity is that of the
radiometric quantity divided by the unit of length, the metre. For example, the dimension of the
-1
unit of radiant intensity Ie is W . sr and the dimension of the unit of spectral radiant intensity
-1 -1
Iel (O) (often written simply as I O are W . sr m , usually reported as mW . sr-1 nm -1 or
PW . sr nm , to provide a more convenient range of numbers for the values reported.
-1 -1
Note: Spectral concentration can also be expressed as a function of frequency or wave

number, but the wavelength function is the one normally chosen to characterise the
spectral distribution of LEDs. The shape of the spectral distribution will change slightly
depending on the bandwidth of the spectroradiometer used. The bandwidth in full width
at half maximum (FWHM) used at the measurement and any correction applied for
zero-bandwidth should be reported when spectral distribution data are presented. The
bandwidth should not be larger than 5 nm.
7.1.2 Normalisation factor and relative spectral distribution

LEDs emit optical radiation over a limited wavelength range given by O1 d Od O2. It is often
helpful to normalise the spectral distribution function and divide it into two parts, an absolute
om
normalisation factor X e0 taken at wavelength O = O0 with the unit of the spectral concentration
Xe0 X O (O O0 ) (35)
.c
and a relative function S X( O)
X O (O)
up
S x ( O) (36)
X e0
ro
called relative spectral distribution, which comes with a unit of unity, but is still associated with
the geometric measurement conditions as defined for the original quantity. From Eq. 36, the
(absolute) spectral distribution can be written as
ng
X O( O) X e 0 S x ( O) (37)
u
is
7.2 Quantities related to spectral distribution

Fig. 11 illustrates the locations of the characteristic wavelengths described in the following
.L
subsections. The shape is typical of that of all LEDs, with zero values outside the wavelength
w
range O1 d O d O2 and one significant maximum in between. Fig 2 shows typical spectral
distributions for a representative selection of the various LEDs currently available
w
commercially.
w
7.2.1 Peak wavelength

The wavelength at the maximum of the spectral distribution is known as the peak wavelength
Op The (absolute) spectral distribution is usually normalised at this wavelength rather than at
an arbitrary wavelength, to give a relative spectral distribution with a maximum value of unity.
7.2.2 Spectral bandwidth at half intensity levels

The spectral bandwidth at half intensity level ' O0,5 is calculated from the two wavelengths O'0,5
and O"0,5 on either side of Op , where peak intensity drops to 50 %:
' O0,5 = O" 0,5 - O'0,5 (38)
Note: In some applications also the ' O0,1 value is used (see Fig. 11), the bandwidth between
those two wavelengths where the intensity drops to one tenth of the maximum.
7.2.3 Centre wavelength of half intensity bandwidth

The wavelength mid-way between the two limiting wavelengths O'0,5 and O"0,5 of the spectral
bandwidth at the 50 % level is specified as O0,5 m . It is calculated from
24
--`,,```,,,,````-`-`,,`,,`,`,,`---

CIE 127:2007
1
O0,5 m O'0,5 O" 0,5 (39)
2
0,8
0.8
0,6
0.6
S(O) )
S(
O
0,4
0.4
0,2
0.2
00
Op Oc O
om
' 'O
O0,50.5 (nm)
(nm)
' 'O
O0,10.1
.c
Fig. 11. Typical relative spectral distribution of a LED showing the location of the
up
characteristic wavelengths and wavelength intervals.
ro
7.2.4 Centroid wavelength

ng
The centroid wavelength Oc of the spectral distribution, which is calculated as the "centre of
gravity wavelength" according to the equation
u
O2 O2
is
Oc O S X (O) d O S X (O) d O (40)

3 3
.L
O1 O
1
w
It should be noted that unlike the other characterizing wavelengths defined here, the
centroid wavelength, when calculated for the types of spectral distribution typical of many LEDs,
w
may be strongly affected by the very small values of the relative spectral distribution at the
diminishing tails of the curve, where measurement uncertainty is increased due to the influence
w
of stray radiation, noise effects or amplifier offsets.
7.3 Colorimetric quantities determined from the spectral distribution -

-`
,
`,
,
,`
,
The colour of the light emitted by a LED may be specified in terms of its chromaticity `
,,
`
`-
-
coordinates and these are best obtained by calculation from the spectral power distribution. ``
`
`
,,
Two alternative quantities also sometimes used to characterize the colour of single-colour ,
,`
`
`,
LEDs are dominant wavelength and purity, they can not be used for white LEDs. They can ,
`
--
be used to provide a quantitative measure of the hue and saturation of the colour and can
be calculated from the chromaticity co-ordinates as explained below. Fig. 12 illustrates the
concepts of dominant wavelength and excitation purity. For further information on
colorimetric concepts and calculations, see reference (CIE, 2004). White LEDs are
characterized by their correlated colour temperature.
25
CIE 127:2007
0,9
520
0,8 540
0,7
560
0,6
Od
500
0,5 DD 580
yy 0,4 C
C 600
620
0,3 N
N
700
0,2
0, 480
460
0
380
om
0 0,2 0,4 0,6 0,8
x
.c
Fig. 12. CIE 1931 chromaticity diagram showing distances and intersections for dominant
wavelength and excitation purity calculations.
up
7.3.1 Dominant wavelength
ro
The dominant wavelength Od of a colour stimulus is defined as follows:

Wavelength of the monochromatic stimulus that, when additively mixed in suitable
ng
proportions with the specified achromatic stimulus, matches the colour stimulus considered.
For characterising LEDs, the reference achromatic stimulus should be an equi-energy
u
spectrum, a stimulus whose spectral concentration of power as a function of wavelength is

constant (sometimes known as illuminant E) and which has the chromaticity coordinates
is
x E = 0,3333, y E = 0,3333.
.L
7.3.2 Purity
w
For characterising purity of LED emission, the term excitation purity p e is used. This is defined
w
as follows:
quantity defined by the ratio NC/ND of two collinear distances on the chromaticity diagram of
w
the CIE 1931 standard colorimetric system (the 1964 diagram gives slightly different values
for lower saturation), the first distance being that between the point C representing the colour
stimulus considered and the point N representing the specified achromatic stimulus; the
second distance is that between the point N and the point D on the spectrum locus at the
dominant wavelength of the colour stimulus considered
The definition leads to the following expressions:
y yn x xn
pe or pe (41)
yd yn xd xn
where ( x, y), ( x n , yn ), ( x d , y d ) are the x, y chromaticity co-ordinates of the points C, N, and D,

respectively.
Note: The value of excitation purity is unity if the chromaticity under test is located on the
spectrum locus. The value is zero if the chromaticity under test has the same
chromaticity co-ordinates as the specified achromatic stimulus.
26 --`,,```,,,,````-`-`,,`,,`,`,,`---

CIE 127:2007
7.4 Spectral measurement of LEDs

For general guidance on spectral measurement of light sources, refer to CIE 63 (CIE, 1984b).
This section describes requirements of spectroradiometers specifically for LEDs. The spectral
distribution and colour of LEDs can be measured with a spectroradiometer in four different
modes: 1) irradiance mode, 2) total flux mode, 3) partial flux mode and 4) radiance mode. In
irradiance mode, the spectral distribution and colour of a test LED are measured in one
direction, whereas, in total flux mode, they are measured as an average for all directions. The
partial flux mode is in between. The radiance mode measures the spectral radiance of the
LED surface, using an imaging optic with the photometer. The irradiance mode can be used
for most typical single-colour LEDs, which have spatially uniform distribution of colour. White
LEDs tend to have non-uniform spatial distribution of colour, in which case the total flux mode
or partial flux mode is recommended to measure the average colour.
7.4.1 Irradiance mode

The radiation from a test LED propagating Integrating
sphere
in one direction within a given small solid
angle is measured. Fig. 13 shows some
examples of the input geometry for the
om
irradiance mode arrangement. The
Spectroradiometer
spectro-radiometer is calibrated with a Spectral irradiance
standard lamp
spectral irradiance standard lamp
(normally, a quartz-halogen tungsten
.c
lamp). It is important that the radiation Fiber-optic
from the standard lamp and test LED is
up input
introduced to the spectroradiometer
(entrance slit or fiber input) exactly in the Transmitting
ro
diffuser
same spatial and angular distribution, and
with the same polarization condition.
(Note that tungsten lamps are slightly
ng
polarized.) To achieve this, a small

Reflecting
integrating sphere (e.g., 50 mm diameter) Diffuser
u
or a diffuser is normally needed for the

input optics. Such input optics should
is
provide spatially and angularly uniform

.L
illumination for the dispersive element, Radiance-mode

independent of the source size and its spectroradiometer
w
intensity distribution. Therefore, the LED

and standard lamp can be different in
w
size, and can be placed at different Fig. 13 . Examples of input geometry for
positions for similar signal level. irradiance mode spectroradiometer.
w
Among the examples given in Fig. 13, the use of integrating sphere (not limited to the
geometry shown in the figure) is most recommended for any spectroradiometer, though the
sensitivity is the lowest. Reflecting diffusers may be used when higher sensitivity is required.
When using a diffuser (especially when combined with fiber optics), caution should be paid to
the spatial nonuniformity of irradiance responsivity over the diffuser surface. It may not be
suitable for Averaged LED Intensity measurement if the spatial uniformity of responsivity is poor
(see Section 3.2). A diffuser exhibiting poor spatial uniformity would work only if it is illuminated
uniformly by the source being measured. Use of a fiber-optic input without diffuser should be
avoided even for simple colour measurement, since the measurement would be sensitive to
incident angle, and thus, the size of the source. For colour measurement of typical LEDs, the
collection angle should be chosen to be 10 ° (+/- 5 ° ) or less. If enough signal is obtained, use of
CIE Condition B or CIE Condition A geometry (see Section 4.3 Averaged LED Intensity) is
recommended. For Averaged LED Intensity measurement, the size of the aperture should be
chosen to satisfy the CIE Condition A or CIE Condition B geometry.
7.4.1.1 Measurement of Averaged LED Intensity

If the spectroradiometer with the irradiance mode input optics is arranged in CIE-B or CIE-A
geometry, and is calibrated for absolute spectral irradiance, a test LED can be measured for
27
--`,,```,,,,````-`-`,,`,,`,`,,`---

CIE 127:2007
-2 -1
absolute spectral irradiance E( O) [Wm nm ]. From this and the distance d [m] of the LED
position (100 mm or 316 mm), the Averaged LED Intensity ILED can be obtained by
2
I LED d Km
3O E (O)V ( O) dO; ( K m = 683 lm/W) (42)
When this method is adopted, the results should be checked by measuring standard
LEDs of different colours. If the agreement is within the stated uncertainty of measurement, the
results are valid with no correction. Disagreement by more than combined uncertainties at
coverage factor k = 2 may indicate incomplete uncertainty budgets and should be investigated
(ISO, 1993). If desired, the spectral calibration can be taken as only relative, and the absolute
scale should be given from the standard LEDs.
7.4.2 Total flux mode

The (spatially) averaged spectral distribution of
the test LED is measured. This mode is Spectroradiometer
commonly used to measure the spectral
distribution and colour of discharge lamps for
general illumination, since they tend to have
spatially non uniform distribution of colour. This
om
mode, if calibrated in absolute units, will
cosine- baffle
measure total spectral radiant flux (unit: W/nm)
of the source. An example of the total flux mode corrected
.c
geometry is shown in Fig. 14. The sphere is the
same geometry as one for total luminous flux Test LED
up
given in Fig. 9(a) with its photometer replaced
by a spectroradiometer. The standard LED and
auxiliary LED are also replaced by a standard
ro
lamp and auxiliary lamp. The spectral flux is --

-
`
spatially integrated over the entire solid angle ,,

`
,
`
ng
,
`,
(4 S sr). Such a system (spectroradiometer + ,

,
-`
`
-
sphere) is calibrated with a total spectral radiant `

`
``
Total spectral radiant flux ,
flux standard lamp (normally, a tungsten lamp). Auxiliary lamp ,

,,
u
standard lamp `
`
`
,
,`
-
-
The self-absorption will be a function of

is
wavelength, so, even if the measurement is for Fig. 14. An example of the geometry for
total flux mode spectroradiometer.
.L
relative spectral distribution, a self-absorption

correction should be applied.
w
7.4.2.1 Measurement of total luminous flux

w
If the spectroradiometer in total flux mode is calibrated for absolute total spectral radiant flux,
w
a test LED can be measured for absolute total spectral radiant flux ) (O) [W/nm]. From this,
the total luminous flux ) v [lm] can be obtained by
) Km OV O d O (K m = 683 lm/W) (43)

v
3O)
For such absolute measurement, the self-absorption correction (spectrally) is critical,
since the size and shape of the standard lamp is very different from LEDs.
When this method is adopted, the results should be verified by measuring standard
LEDs of different colours. If the agreement is within the stated uncertainty of measurement, the
results are valid with no correction. Disagreement by more than combined uncertainties at k=2
may indicate incomplete uncertainty budgets and should be investigated. If desired, the spectral
calibration can be taken as only relative, and the absolute scale can be given from the standard
LEDs. In general, the calibration path that leads to lowest measurement uncertainties should be
adopted as the primary calibration procedure.
7.4.3 Partial flux mode

A spectroradiometer can be used also for partial flux mode. Fig. 15 shows an example of
such geometry. This is the same sphere geometry as the one for Partial LED Flux (Fig. 10),
28
CIE 127:2007
with the photometer head replaced by the spectroradiometer, and the auxiliary LED replaced
by the auxiliary lamp. The same recommendations on the construction of the sphere given in
Section 6.2.2.2 apply, except that the requirement for the aperture is not critical if only relative
spectral distribution is measured with this arrangement.
Spectroradiometer
dd baffle cosi ne-

cosine-
corrected
correc te d
Test LED
Spectral irradiance
standard lamp
om
Auxiliary lamp
.c
up
Fig. 15. An example of the geometry for partial flux mode spectroradiometer.
The integrating sphere system with the spectroradiometer can be calibrated with a
ro
spectral irradiance standard lamp. The standard lamp is placed outside the sphere at the
distance d from the aperture, at the same distance where the spectral irradiance is calibrated.
ng
(For example, a 1000 W FEL type lamp is normally calibrated at 50 cm to 70 cm distance.)

2
If the area of the aperture A [m ] is known, the spectral radiant flux ) ext ( O) [W/nm]
-2 -1
u
introduced into the sphere is calculated from the spectral irradiance E(O) [Wm nm ] by
is
) ext ( O) k a A E ( O) (44)
.L
where k a is the correction factor for average irradiance over the sphere opening with respect
to the small area over which the irradiance from the lamp is calibrated. Spectral irradiance
w
standard lamps are normally calibrated for much smaller collection angle (so, much smaller
--`,,```,,,,````-`-`,,`,,`,`,,`---
area). k a can be obtained by spatially mapping the irradiance distribution of the lamp over the
w
area corresponding to the sphere opening. For only relative spectral measurement (e.g., only
for colour), k a and A can be practically ignored.
w
7.4.3.1 Measurement of Partial LED Flux

If the spectroradiometer in partial flux mode is calibrated against the absolute spectral radiant
flux ) ext (O), a test LED can be measured for partial spectral radiant flux ) e, LED, x (O) [W/nm],
from which Partial LED Flux ) LED, x [lm] is obtained by
) Km
LED, x
3O) e, LED.x (O)V (O) d O; ( K m = 683 lm/W) (45)
If the test LED has a mount which is not neglectable and/or placed closer to the
opening, the self-absorption measurement using an auxiliary lamp should be performed and a
correction applied. In some cases self-absorption can be ignored if the test LED and mount is
small and placed far from the opening. When this method is adopted, the results should be
verified by measuring standard LEDs of different colours. If the agreement is within the stated
uncertainty of measurement, the results are valid with no correction. Disagreement by more
than combined uncertainties at k = 2 may indicate incomplete uncertainty budgets and should
be investigated. If desired, the spectral calibration can be taken as only relative, and the
absolute scale can be given from the standard LEDs. In general, the calibration path that leads
to lowest measurement uncertainties should be adopted as the primary calibration procedure.
29
CIE 127:2007
7.4.4 Consideration for bandwidth and scanning interval

When a spectral measurement is made, errors will occur due to the bandwidth of the
spectroradiometer as well as the scanning interval (CIE, 1984b). These errors are more
prominent with LEDs, which have narrow-band emissions. The bandwidth of a
spectroradiometer has an effect of broadening the measured spectra, which will result in
errors in colour quantities obtained. For example, a 10 nm bandwidth (FWHM, triangular
shape) would cause an error of ~0,003 in u’v’ chromaticity for a red and blue LED, and
~0,002 for green LED. Such errors due to the bandwidth of a spectrometer are proportional to
the square of the increase of bandwidth. With a 5 nm bandwidth, the errors in u’
v’
chromaticity will be less than 0,001 for LEDs of any colour. For practical LED measurements,
a bandwidth of 5 nm or less is acceptable and recommended. Bandwidths of larger than 5 nm
are generally not recommended for LED measurements, but might be used with appropriate
bandpass correction (see Section 7.4.4.1).
The scanning interval also causes errors though much smaller than the errors due to
bandpass. For example, the error due to a 10 nm data interval (with 0 nm bandwidth) for typical
LEDs are less than 0,0005 in u’， v’ . The errors for a 5 nm data interval are negligible (<0,00001 in
u ’ ,v ’ ). Though it is a common practice to match the bandwidth and scanning interval for
colorimetry of discharge lamps, the mismatch hardly affects the errors in chromaticity of LEDs.
om
Rather, smaller bandpass is important for a given scanning interval. The scanning interval,
however, is important to obtain such quantities as peak wavelength and spectral width (spectral
shape) of LEDs. If only chromaticity is to be measured, a scanning interval of 5 nm or less is well
--`,,```,,,,````-`-`,,`,,`,`,,`---
acceptable. For measuring peak wavelength and spectral width, an interval of 2,5 nm or less is
.c
recommended. Over-sampling (e.g., 2,5 nm interval for 5 nm bandwidth) is advantageous for
colour measurement, since, for a given bandwidth, it will reduce colorimetric errors due to
up
random noise as well as due to sampling errors
While the bandwidth and scanning interval are fixed and not changeable in many
ro
commercial instruments (e.g., diode-array type), these parameters are selectable in most
mechanical-scanning type instruments. Smaller bandpass is desirable for less errors but the
ng
signal will be less. Smaller interval is desirable but measurement takes more time, so these
parameters are set depending on the intensity of LEDs and the uncertainty required.
u
7.4.4.1 Bandpass correction

is
For highest accuracy applications, or if the bandwidth is more than 5 nm (but no more than
.L
10 nm), it is recommended that a bandpass correction be applied. The Stearns and Stearns
method (S-S method) (Stearns and Stearns, 1988) is very simple to apply, and very effective.
w
To apply the S-S method, it is required that the bandpass must be a triangular function, and
the bandwidth ' O0,5 and scanning interval Ostep must be matched ( ' O0,5 = n ' Ostep ; n: integer).
w
For example, for 5 nm bandwidth and 5 nm scanning interval, the corrected spectral value Si’
is recalculated from the original values at neighbouring 5 points by
w
S i’ =S(i-2 - 12 Si-1 + 120 S i - 12 Si+1 + S i+2 ) / 98 (46)

where the values of S i-2, S i-1 , Si , S i+1 and S i+2 correspond to, for example, the values at
440 nm, 445 nm, 450 nm, 455 nm, and 460 nm. The bandpass errors will be mostly removed
if the bandpass requirements are physically satisfied accurately. If the bandwidth is 5 nm and
the scanning interval is 2,5 nm, the original values of every other 5 points around the
wavelength are used for the calculation.
Note: In real spectroradiometers, their bandwidth is not perfectly triangular shape and not
perfectly constant over the visible region. Therefore, the results of correction will not be
as perfect as predicted by simulation. Some residual errors should be taken into
account even after the correction. Improved methods that can apply to non-triangular
bandpass and do not require that bandwidth and scanning interval to be matched, have
also become available (Ohno, to be published; Gardner, to be published).
7.4.5 Other uncertainty components

Other than bandwidth and scanning interval effects, there are many other uncertainty
components in spectral measurements of light sources, such as wavelength error, noise,
stray light, fluorescence, linearity of detector, uncertainty in spectral standards, etc. Noise
30
CIE 127:2007
(dynamic range) and stray light are particularly critical for LED measurements because typical
LEDs have emission only in a part of the visible region. If stray light signal falls on the region
where the LED has no emission, the effect in colour will be large. Some commercial
instruments do not allow negative values, of spectral distribution (due to noise) which is not
acceptable for LED colour measurement. In the region where a LED has no emission, if
negative noise is all truncated to zero, the remaining positive noise would result in similar
effects to stray light and can cause significant error in chromaticity. In the case of UV-LEDs,
the measurement uncertainty due to fluorescence is becoming more significant.
Uncertainties in spectral values or in wavelength uncertainties will propagate into the
uncertainties in colour quantities (such as chromaticity coordinates, correlated colour
temperature, dominant wavelength) by statistical methods (ISO, 1993). Practical methods to
calculate colour uncertainties of light sources (including LEDs) from the uncertainty of spectral
values or uncertainty in wavelength are available (Ohno, 2001; Gardner, 2000).
Other details of uncertainties in spectral measurements are common with measurement
of general light sources, and are beyond the scope of this report. These are covered in Ref.
(CIE, 1984b).
om
8. REFERENCES
CIE, 1983. CIE 18.2-1983. The Basis of Physical Photometry, 1983.
CIE, 1984a. CIE 64-1984. Determination of the Spectral R esponsivity of Optical R adiation
.c
Detectors, 1984. up
CIE, 1984b. CIE 63-1984. The Spectroradiometric Measurement of Light Sources, 1984.
CIE, 1987a. CIE 17.4-1987. International Lighting Vocabulary, 1987.
ro
CIE, 1987b. CIE 69-1987. Methods of Characterising Illuminance Meters and Luminance
Meters, 1987.
ng
CIE, 1987c. CIE 70-1987. The Measurement of Absolute Luminous Intensity Distributions,
1987.
u
CIE, 1989. CIE 84-1989. Measurement of Luminous Flux, 1989.

is
CIE, 2004. CIE 15:2004. Colorimetry , 2004.

.L
CIE, to be published a. Photometry Using V( O)-corrected Detectors as R eference and

w
Transfer Standards . A report from TC 2-37. To be published.

CIE, to be published b. Determination of Measurement Uncertainties in Photometry . A report
w
from TC 2-43. To be published.

w
CSUTI, P., KRANICZ, B., and SCHANDA, J., to be published. Description of a partial f1 ’ error
index recommended for LED photometry. To be published in: Svetoteckhnika/Light &
Engineering.
GARDNER, J.L., 2000. Uncertainty estimation in colour measurement. Color Res. & Appl. , 25 ,
349-355, 2000.
GARDNER, J.L., to be published. Bandwidth Correction for LED Chromaticity. Color Res. &
Appl., to be published.
GOURE, J.P. and MASSOT, J.N., 1982. Commentaires sur la d é termination de la loi de
luminance L(P,q) d ’ une source semi-conductrice en champ proche. Optical and Quantum -
-
Electronics , 14, 5, 1982. -`

,
,
`
,
`,
,
`
GRUM, F. and BARTELSON, C.J ., 1980. Optical radiation measurements, Vol. 2, Color ,,
`
-
`
-
Measurements . Academic Press, New York, 1980. ``

`
`
,,
,
`,
`
ISO, 1993. Guide to the Expression of Uncertainty in Measurement , 1993. ,`

,
`
--
OHNO, Y., 2001. A Numerical Method for Color Uncertainty. In Proc. CIE Expert Symposium
2001 on Uncertainty Evaluation , CIE x020-2001, 8-11, 2001.
31
CIE 127:2007
OHNO, Y., to be published. A Flexible Bandpass Correction Method for Spectrometers. In

Proc., AIC ’, 05
Granada, Spain, 9-13 May, 2005, to be published.
STEARNS, E.I., and STEARNS, R.E., 1988. An Example of a Method for Correcting
Radiance Data for Bandpass Error. Color Res. & Appl . 13, 257-259, 1988.
WYSZECKI, G. and STILES, W.S., 1982. Color Science, 2nd ed. John Wiley & Sons, New
York, 1982.
--`,,```,,,,````-`-`,,`,,`,`,,`---
om
.c
up
ro
u ng
is
.L
w
w
w
32

High Precision Spectroradiometer Integrating Sphere System

Uploaded by

Copyright:

Available Formats

High Precision Spectroradiometer Integrating Sphere System

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

High Precision Spectroradiometer Integrating Sphere System

Uploaded by

Copyright:

Available Formats

2022/7/7 High Precision Spectroradiometer Integrating Sphere System for LED Light Measurement - LISUN

 Sales@Lisungroup.com |  +8618117273997 Search...  English 

Home  Products  High Precision Spectroradiometer Integrating Sphere System

High P recision Spectroradiometer Integrating Sphere System

Goniophotometer  Get a Quote

LED Test Instruments  Name* Company*

CFL Testing Instruments 

Photometer and Colorimeter  Email* Cell/WhatsApp

EMI and EMC Test Systems 

Electrical Safety Tester 

AC and DC Power Supply 

Object Color and Glossiness Test  61

Electronic Components Test 

Description Video Download

 Sales@Lisungroup.com |  +8618117273997 Measures: Search...  English 

Fluorescent Lamp Test Solutions

NOM Norma Oficial Mexicana

LISUN Model LMS-9000C LMS-9000CUV-VIS LMS-9000CVIS-NIR

Wavelength 350-800nm 200-800nm 350-1050nm

 Sales@Lisungroup.com |  +8618117273997 Search...  English 

Related Successful Case

Turkey-Install the LSG-1800BCCD High

Turkey–Install LSG-1700B Goniophotometer by

Iran-Free installation and training for LPCE-

India-Installation and training for LSG-1700B

India-LISUN engineer visit India to do

Egypt – Installation of LSG-1800B

Goniophotometer | Integrating Sphere | Surge Generator      © Copyright 2003 | Blog | Privacy

Copyright International Commission on Illumination

LA COMMISSION INTERNATIONALE DE L'ECLAIRAGE

DIE INTERNATIONALE BELEUCHTUNGSKOMMISSION

Die Ziele der CIE sind :

Copyright International Commission on Illumination

Copyright International Commission on Illumination

Ce rapport technique a é t é é labor é par le Comit é Technique CIE 2-45 de la Division 2

m?gliche sp?tere ?nderungen zu Rate gezogen werden.

these may not be comprehensive.

ne saurait ê tre exhaustif.

CIE 2007 – All rights reserved

Copyright International Commission on Illumination

1.3 Purpose of the report 1 `,

1.4 Categories of LED measurement 2 ``

1.4.1 Laboratory measurements 2 `

1.4.2 Bulk testing 2 -

2.3.1 Shift of peak wavelength with temperature 8

2.4 Production tolerances 9

3.3 Spectral responsivity of the photometers / radiometers 10

3.3.1 Photometer to measure white LEDs 11

4. QUANTITIES DEFINING SPATIAL RELATIONS 11

4.1 Normalisation factor and relative spatial distribution 11

4.2.1 Luminous intensity 13

6.2.2 Integrating sphere method 20

MESURE DES DIODES ELECTROLUMINESCENTES (LED)

Ce rapport est une actualisation du rapport technique CIE 127-1997 publi

é talons doivent ê tre é talonn é es par un laboratoire national de m é trologie ou un laboratoire

MESSUNG VON LEDS

x Averaged LED Intensity;

x Partial LED Flux.

Please see document for exact definitions.