Characteristic test of general lighting LED light source

1 Overview
The research and development, production and application of LED have developed rapidly in recent years. The invention of the blue LED makes it possible for LEDs to be used in the field of general lighting. With continuous breakthroughs in high-efficiency and high-power LED technology, LEDs will likely become a widely used lighting source in this century.
The Chinese government and the international community have paid great attention to the development of LED. The second division of the International Commission on Illumination (CIE), the Division 2: Measurement of Light and Radiation (Division2: Measurement of Light and Radiation), has held two representative seminars for the measurement of the characteristics of LEDs. , And achieved certain results. However, the physical measurement of LEDs has not yet been fully resolved. In June 2004, CIE held another CIE LED Symposium (CIE LED Expert Symposium) in Japan to further discuss the problem of LED characteristic measurement.
Regarding the characteristics of LEDs, due to the different application fields, a unified understanding is needed. There are obvious differences in the understanding of LEDs between the semiconductor industry and the lighting industry. The characteristics of LEDs used for indication (including display) purposes and the characteristics of LEDs used for lighting purposes There are also differences, the corresponding physical measurement methods and quality assessment methods are also different, the instruments used are also different, and the test conditions of each laboratory are also different.
Due to the above reasons, it often happens that various laboratories give different measurement results to the same LED. This article mainly introduces the characteristic measurement of lighting LEDs, introduces the achievements of CIE in LED measurement, the more generally recognized physical measurement methods of CIE Division 2 (Division2), and the special measuring instruments based on the existing CIE technical documents. A complete physical measurement of a typical LED light source for general lighting is presented.
Characteristic test of general lighting LED light source
2, LED light source for general lighting and its characteristics
General lighting LED, commonly known as "white light" LED. At present, there are two main ways to realize white light LED: use the blue light emitted by the LED to excite the phosphor, and then mix the light with the "yellow light" emitted by the phosphor to obtain white light; second, use three types of red, green, and blue Different LEDs are connected and mixed to obtain white light.
Whether it is a single packaged LED or an LED light source composed of LED strings, it is not a point light source or Lambertian light source that is easy to understand and describe in physics. It is a scattered point light source composed of multiple equivalent point light sources. (Image: both virtual and real images exist, and most of them are virtual images). The specific point light source set corresponding to the LED light source has common characteristics:
(1), the light intensity distribution is different at different distances in the near field;
(2). At a certain angle (such as the normal direction), the near-field luminous intensity is not constant. Therefore, the law that the illuminance is inversely proportional to the square of the distance does not hold;
(3). Due to the position sensitivity of the LED chip and the package lens and the asymmetry of the LED chip, the LED light source has different light intensity distribution curves on different C planes;
(4) The "white light" LED that has entered industrial production at present, its spectral mixing is not uniform, and its color distribution is like a light intensity distribution curve, which changes with distance and angle, especially pcLED, often There is a blue color center with a high color temperature.
3. Measuring methods and instruments for LED light sources for general lighting
It is precisely because of the above-mentioned characteristics of the LED light source for general lighting, its measurement and quality evaluation require specific methods and special equipment to solve it. CIE has formulated some standards for LED measurement, and some standards are still under development. Some standards will become CIE/ISO and CIE/IEC joint standards to unify international LED measurement issues. Although there are only a few scientific and technological workers in China who are engaged in LED measurement, China's LED measurement methods and instruments have been tracking the world's dynamics and have received the attention of CIE. The physical measurement methods and instruments of the main characteristics of the lighting LED are introduced one by one below.
3.1 Measurement of average normal light intensity
Average normal light intensity is a commonly used LED quality indicator. CIE PublicaTIon No.127-1997 gives clear regulations and descriptions. Figure 4 clearly expresses the basic measurement method. Figure 5 is a measuring instrument based on CIE PublicaTIon No.127. This instrument can not only realize CIE CondiTIon A and B at the same time, but also can rotate the angle of the LED under test, so as to measure the average light intensity of the LED under different angles.
3.2 Measurement of average light intensity distribution curve and light intensity distribution curve
Record the illuminance values at different θ angles, and then calculate the average light intensity according to the inverse square law of distance, then the resulting curves are the average light intensity distribution curves under CIE CondiTIon A and B respectively.
Although it has been stated that LEDs do not meet the characteristics of point light sources and do not satisfy the inverse squared distance law, when the distance is large enough, such as: 10 times the specific point light source set (image) corresponding to the LED light source, the distance squared The law of inverse ratio is restored to conformity again. Since the opening angle formed by the detector's aperture is sufficiently small at this time, the word "average" is no longer important, and the word "" can be added, but generally it can be omitted. Figure 6 is the light intensity distribution curve of a 1W lighting LED under CIE Condition A and B respectively.
Special attention should be paid to the choice of the measured plane when testing the average and light intensity distribution curve of the LED. For convenience, this article defines the plane determined by the center of the LED angle as the CO° plane, and the measurement is made on this plane. The distribution of light intensity with the angle θ. The long-distance test (under point light source conditions) may involve a illuminance meter with a higher sensitivity and a wide linear dynamic range. This is a relatively expensive instrument, and the test environment requirements are relatively high. The relevant test equipment is shown in Figure 7.
In addition, because the LED light source is a narrow-band spectrum light source, according to the requirements of CIE Publication No.127, it is necessary to perform the SCF correction of the detector according to formula (1), but experiments show that high-quality V(λ) detection When measuring the LED light source for general lighting, the instrument can get satisfactory results even if it is not corrected, and its uncertainty can be ±3% or even better.
3.3 Measurement of luminous flux and color
Due to the unevenness of the spectrum of light emitted by the LED light source for general lighting, sufficient light mixing should be performed when measuring the color characteristics of the LED, otherwise the measurement results cannot be unified. Therefore, for the LED light source for general lighting, the measured Strictly speaking, the color should be the total average color
It is more reasonable to measure the color while measuring the luminous flux in the integrating sphere. China has its own unique features in LED luminous flux and color measurement. In addition to complying with CIE’s general regulations on light and color measurement publications, China also organically combines integration and spectroscopy when measuring luminous flux and color (referred to as " Combination method”), this creative technology has been applied for.
This system uses a Si-V (λ) detector to measure luminous flux in accordance with the requirements of CIE Publication No.84-1989. This method has a good wide linear dynamic range for measuring luminous flux. According to CIE Publication No.63-1984, the spectral power distribution of the lighting LED was measured by a spectrophotometer composed of a monochromator and PMT, and then the SCF correction method described in CIE Publication No.127-1997 was used to automatically calibrate the Si- The V(λ) detector measures the luminous flux system. The correction factor is:
∫780380PTe(λ)·V(λ)·dV ∫ 780380Pse(λ)·R(λ)·dλ
SFC = ——------——————————? ————---------————————
∫780380PTe(λ)·R(λ)·dV ∫ 780380Pse(λ)·V(λ)·dλ
  (1)
In the formula: PTe(λ) is the relative spectral power of the light source under test, and the article should be the LED light source for general lighting;
Pse(λ) is the relative spectral power of a standard lamp (usually an incandescent lamp or a tungsten halogen lamp);
R(λ) is the relative spectral sensitivity of the Si-V(λ) detector, including the equivalent spectral transmittance of the integrating sphere;
V(λ) is the optical efficiency function of CIE spectrum;
SCF is the spectral mismatch correction coefficient.
In this system, the SCF calibration is very easy, so the luminous flux measurement has excellent dynamic range and measurement accuracy. It is possible to achieve high-precision calibration of the system with a single standard lamp. At the same time, the system has the same out-of-spectrum error in the small dynamic range (the harsh conditions that require the lamp under test to be basically close to the standard light flux) as the ordinary spectroscopy method. As shown in Figure 9, the linear dynamic range of the combined method is 5 higher than that of the ordinary spectroscopy method. Because in the ordinary spectroscopy, at least 3 orders of dynamic range are left to scan the relative power distribution of the spectrum, so the standard The accuracy line of the ordinary spectroscopy with a good lamp calibration must be only 1-2 orders of magnitude, which is extremely narrow.
Due to the increasing power of lighting LEDs, both individual LEDs and LED strings need to be measured (requires a large dynamic range of the instrument), and the spectral distribution of LEDs is relatively special. Therefore, the combined method has the advantages of measuring the luminous flux and color of LED light sources for general lighting. More obvious.
4. Discussion
4.1 About CIE Condition A and B
Generally, the measurement range of the commonly used illuminance meter is 0.1 lx, and the corresponding light intensity at 100mm and 316mm are 1mcd and 10mcd respectively, while the light intensity of the previous low-power LED is generally between 1-2000mcd. Therefore, the formulation of Condition A and Condition B by CIE Publication No.127-1997 is very conducive to the simple measurement of LED normal light intensity.
Nowadays, lighting high-power LEDs are continuously being developed and applied. The power of a single LED for lighting has been developed from tens of mW to several W, which has increased by two orders of magnitude, and the luminous efficiency is also increasing. Therefore, the average normal light intensity of the LED is expected to increase 2-3 orders of magnitude, which means that 1-1000cd LEDs will be more common. If this is the case, it is recommended to change the measuring distance in CIE Condition A and B to 1m or 3.16m. This change not only allows ordinary illuminance meters to easily measure the lighting LED, but also measures the aperture (sensitive area) of the contrast meter. There is no need to make strict restrictions. The measured light intensity is no longer the near-field light intensity. Therefore, the inverse square of the distance law can be used again. The measured light intensity can be changed from "average" to "normal light intensity." The dependence on test equipment is further weakened, which is more conducive to the international quality assessment of LEDs in industry and commerce.
4.2 Calibration and standard lamp of lighting LED measuring instrument
Use LED as the secondary standard light source to calibrate the instrument, or to calibrate the instrument with a standard incandescent lamp, is a question that has always been discussed. The author believes that laboratories with better conditions should use standard incandescent lamps to calibrate the instrument, especially in laboratories that can test the SCF value, because standard incandescent lamps have better stability and reproducibility. . However, for factories and workshops with poor conditions, it is a very practical method to calibrate the instrument with the same type of LED that has been tested, so sometimes it is necessary to pay attention to the geometric conditions of the LED and the ambient temperature.
4.3 Positioning reference when measuring LED light intensity
When measuring the light intensity and light intensity distribution, the positioning reference of the LED light source is very important. Although CIE No.127-1997 has clearly used the mechanical axis and the front vertex as the positioning reference, the actual operation is due to the small size of the LED light source. It may still lead to larger measurement errors due to positioning errors. Therefore, whether it is better to use the peak light intensity axis as the reference positioning reference is a question worthy of further discussion.
4.4 Environmental conditions for LED test
In the LED test, in addition to the constant current mode of power supply, geometric conditions and ambient temperature conditions are extremely important. If you do not pay attention to the plane to be measured when measuring the light intensity distribution curve, or if it is the same plane but there is a problem with the alignment of the mechanical axis, then an error of ±20% or more is completely possible.
In addition, the LED light source is a highly temperature-dependent light source. The increase in temperature may cause the red shift of the peak wavelength of the LED light and the rapid attenuation of the light output. Therefore, maintain effective control of the ambient temperature in the LED test (such as using a constant temperature Integrating sphere), or monitoring the temperature of the LED itself is very meaningful. This is also one of the important conditions to ensure that the measurement results can be reproduced.
  5 Conclusion
Using the above measurement method and instrument to repeatedly test the light source of LED for general lighting can achieve good stability and reproducibility. This shows that: under the framework of the existing CIE documents, according to the unique characteristics of LEDs, some CIE measurement methods and conditions are revised and adjusted in a small amount, and the corresponding CIE recommended methods are further formulated. The problem of measuring the characteristics of LEDs for general lighting is completely It can be solved.