Standard techniques for optical radiation measurements, in general, are developed by the Commission Internationale De L'Eclairage (CIE), (the International Commission on Illumination). The issues of LED measurements were raised several years ago, and in 1997, CIE issued Publication No. 127, "Measurement of LEDs", which established guidelines for measuring luminous intensity of LEDs.
Luminous intensity is a quantity that strictly describes a point source and most LEDs are not point sources at close distances; therefore the measured luminous intensity varies depending on the distance and size of the photometer aperture. To solve this problem, Publication 127 introduced standardized geometries: the size of the photometer aperture is 1 cm2 (circular), and the distance between the LED and the photometer is 316 mm or 100 mm (see figure 1).
These fixed geometries have made the measurements reproducible and comparable. The quantity measured is not a true luminous intensity. LED spatial distributions change radically with angle. Therefore, the measurement across the photometer aperture averages the luminous intensity, so it is named Averaged LED Intensity. The two standardized geometries are designated as CIE Condition A (d = 316 mm) and Condition B (d = 100 mm), and the symbols ILED A and ILED B are used. This recommendation has proved successful in solving the problem in luminous intensity comparisons and is widely used.
The CIE has seven Divisions, and the area of optical radiation measurements is covered by Division 2 (Physical Measurement of Light and Radiation). Standardization work is done by Technical Committees (TCs) assigned to each task. Currently, as many as four TCs out of 28 TCs in Division 2 are related to LED measurements, which indicates the high level of interest in LED measurement issues.
Revisions to CIE 127
TC2-45 is preparing a revision of CIE 127. The main reason is that CIE 127 was focused on luminous intensity measurements and did not cover sufficiently the measurement of total luminous flux and color of LEDs, which are now becoming very important for solid state lighting and signaling applications.
Another reason is that LED technologies are changing rapidly and some measurement methods must be updated for varieties of new LEDs. There were also some possible problems in the measurement geometries for total flux measurement described in CIE 127.
The revision is to include improved integrating sphere geometries for total luminous flux measurements. In addition to total luminous flux (integration of flux over 4π steradian), Partial LED Flux (flux emitted within a given cone angle) is being considered (see figure 2).
This property is defined as the flux leaving the LED and propagating within a given cone angle, centered from the LED's mechanical axis. Any flux emitted in the directions other than in this cone angle is ignored. Such a quantity is considered because, in many applications, the flux emitted only in the forward direction is used, and such partial flux (without backward emission) needs to be evaluated.
Recently, spectroradiometers are increasingly being used for LED measurements. The revision is also to give recommendations on using spectroradiometers not only for measurement of color but also for luminous intensity and luminous flux of LEDs; in other words, to use a spectroradiometer as a "perfect" V(λ)-corrected photometer, which does not cause spectral mismatch errors (theoretically).
TC2-46 is preparing a CIE/ISO standard on the measurement of Average LED intensity, which will bring the luminous intensity part of the recommendation in CIE 127 to a higher status of joint international standards by CIE and ISO.
Clusters and arrays
Many questions and concerns exist in the measurement of LED clusters and arrays. Two groups are active in preparing recommendations for these. CIE TC2-50 has recently started to prepare a technical report for measurement of the optical properties of visible LED clusters and arrays, to give recommendations for definitions and measurement methods and conditions for various clusters and arrays of LEDs including static displays and signs.
Also, within the Illuminating Engineering Society of North America (IESNA), the Testing Procedures Committee has Project 78 - Guide for Measurement of LEDs. The purpose of this project is to develop a guide for appropriate methods and equipment to measure the output of LED fixtures for lighting and signaling purposes.
Visual aspects and CRI
CIE is also investigating the visual aspects of LEDs; this work is done in CIE's Division 1 (Vision and Color). Technical committee 1-62 "Colour Rendering of White LED Light Sources" is investigating the color rendering properties of white LED light sources by visual experiments and to test the applicability of the CIE color rendering index (CRI) to white LEDs.
The CRI was originally developed nearly 40 years ago, and even after a few revisions, it still uses outdated formulae and methods, and it is known to have deficiencies. The problem is prominent with narrow-band sources, and the use of CRI for RGB white LEDs or multi-chip white LEDs can mislead design directions. The problems are under investigation at NIST and other institutes, and improvement of CRI or different new methods, particularly for solid-state lighting sources, are being studied.
On the side of photobiological safety, two documents exist; IEC 60825-1-2001 "Safety of laser products part 1: equipment classification and user's guide" and CIE S009/E: 2002 "Photobiological safety of lamps and lamp systems". Both of the documents include LEDs in their scope. The problem is that LEDs are treated as lasers in IEC 60825.
The photobiological area is covered by CIE Division 6 (Photobiology). TC 6-55 "Photobiological Safety of LEDs" is to report on the differing methods of assessing the photobiological safety of LEDs. This entails a review and report on the known effects from a physiological standpoint and a determination of the dose relationships that pose a potential risk for eye injury from excessive irradiation.
One aspect that makes the evaluation difficult is the measurement of radiance of LEDs, which is the quantity that relates to the level of injury. However, due to small size and very directional characteristics of LEDs, measurement of radiance is very difficult and a large uncertainty is involved. Also, there is an effect that, if the size of the source is very small, the effective source will be larger (thus effective luminance will reduce) due to small eye movement.
CIE TC 2-58 "Measurement of LED Radiance and Luminance" was formed in 2003 to prepare a Technical Report setting out recommendations for measurement of the radiance and luminance of LEDs, taking particular account of the specific requirements for evaluation of photobiological safety.
The CIE Technical Committees consist of experts in the subject area as members from various countries, assigned by the chairperson or through national committees of CIE, such as CIE-USA for example. The TC drafts are often discussed by e-mail as well as at occasional TC meetings.
CIE also holds symposia. There were three symposia on LED measurements recently; the first one in 1997 in Vienna, the second one in 2001 in Gaithersburg, Maryland, and third one in 2004 in Tokyo, Japan on a broader topic: "Symposium on LED Light Sources". Some of the presentations at these Symposia are often considered and discussed in TC meetings.
For any questions on these CIE activities, please contact Yoshi Ohno, Secretary of CIE Division 2.