Near-field optical distribution test of LED light source

LED semiconductor lighting network LED optical design is a very important part of lighting products. Before the advent of the near-field goniophotometer, the model of the primary light source can only be "theoretical modeling" by the optical design engineer's ability and simulation software. . Because there are differences between theoretical modeling and actual models, there are two major problems in this type of modeling. First, most of the lenses need to be re-opened, which is expensive; second, product development takes a long time. Therefore, the advent of the near-field goniophotometer is the best gift for optical design engineers. In recent years, as our optical quality requirements for LED products have become higher and higher, the demand for optical field near-field optical distribution models has continued to grow. Both LED light source manufacturers and lens manufacturers require optical optics. Distribution model. So, what is the near-field optical distribution model of the light source? How to use the near-field goniophotometer to obtain the near-field model? What types of files can be generated by near field testing and how can I use the various documents generated? In addition, for unconventional lamp beads, such as high-power modules, bare crystal, 5-sided CSP and other light sources, how can it be handled? This article will answer you one by one.

1. Near-field optical distribution model and test principle of light source

At present, in the LED optical design process, two models are generally used to simulate the light source, namely "light source far field model" and "light source near field model". Before you understand the near-field model of the light source, let's briefly introduce the familiar far-field model of the light source.

The far-field model of the light source is to treat the light source as a point source, and all the light rays are emitted from the same point. Generally, the light output from the point source is isotropic. The far-field model of the source is measured by a far-field distribution photometer, which typically includes a mechanical structure (turntable) and a photometric detector for supporting and positioning the source under test. According to the requirements of CIE70, the distance between the light source and the detector needs to be far enough when measuring (generally, the measurement distance should be at least 5 times of the maximum luminous surface of the LED light source), and the light source can be regarded as a point light source.

For LED light sources, especially white light sources, the brightness and color of the surface are not evenly distributed due to the effects of electrode design, chip structure, and phosphor coating. As shown in Figure 1, it can be seen that the brightness of each chip and the color distribution of the surface of the light source are not completely identical. The secondary optical design of the LED light source is relatively rough, and the information of the light source obtained by the far-field model is relatively rough, which cannot accurately reflect the brightness and chromaticity spatial distribution difference of the LED light source surface, and it is difficult to achieve accurate secondary optical design for the light source. Therefore, accurate measurement of the luminescence model of the source itself is critical to the accuracy of the optical design and simulation results.

a) true color map b) pseudo color map

Figure 1 White LED light source surface brightness distribution

That is to say, the most essential difference from the far-field model of the light source is that the far-field model of the light source considers the light source as a point source, while the near-field model of the source considers the source as a complex surface source. The shape of the light source is represented by a plane, and all light rays are emitted from the surface of the light source. The near-field model is closer to the actual light-emitting condition of the LED light source. The measurement can obtain the brightness and chromaticity values ​​of each point in the measured plane, and provide more accurate and detailed data for the optical design of the LED light source.

The near-field model of the source can be measured by a near-field distribution photometer. As shown in Figure 2 a), the near-field distribution photometer consists of a distribution photometer and an imaging luminance meter, which replaces the photometric detector in the distribution photometer. The imaging luminance meter uses a two-dimensional optical receiving component (such as a CCD), and one sampling can measure the brightness value of each point in the measured plane. The imaging luminance meter in the near-field distribution photometer directly receives the light beam of the LED light source facing the LED light source under test. The light beams emitted by the measured light source all have measurable distance-independent brightness values. By measuring the brightness values ​​of the respective light-emitting points on the surface of the tested LED light source in all directions in the space, the ray tracing method can accurately obtain each of the LED light sources. A luminosity parameter such as a illuminance distribution, a spatial light intensity distribution, and a total luminous flux, and is independent of the test distance, direction, or radius of curvature of the LED surface. If the chromaticity information is to be measured, the imaging illuminance meter is replaced by an imaging chromaticity meter to obtain the spatial chromaticity distribution of the LED light source.

As shown in Figure 2 b), the light source can be rotated around its own mechanical axis during the measurement process. The imaging light/colorimeter shoots the light source image from various angles of the space. The measurement results of each specified angle contain brightness and color information, and the light source is constructed. A three-dimensional image of the brightness and chrominance output. At the end of the measurement, the measurement software integrates these images into a near-field model that describes the brightness and color distribution of the source and gives it in the form of light intensity. The intensity I (x, y, z, θ, φ) is the position (x, y). , z) and the function of the angle (θ, φ). This function also contains color coordinate values ​​or spectra if color and spectral measurements are taken. The near-field model of the source generates a set of rays for optical design and far field distribution of the extrapolated source.

Figure 2 a) Schematic diagram of the structure of the near-field optical distribution photometer Figure 2 b) Schematic diagram of the near-field optical distribution photometer

2. Type and application of near-field model of light source

In general, the near-field model of the source can be divided into the following categories:

a. Contains only the brightness or radiance distribution information of the light source, without color information

Basically all simulation software supports this type of optical file, but because there is no color information, the color of the light source cannot be accurately simulated. In general, such near-field models are suitable for testing monochromatic (brightness) or invisible (radiance) sources. The SIG 400 from the Hong Kong University of Science and Technology's Foshan LED Center can test near-field models of light sources with wavelengths ranging from 350 nm to 1000 nm, including UV and IR invisible light.

Figure 3 Near-field model of the light source with only luminance information opens the interface in professional analysis software

b. Distribution information including luminance and chrominance (tristimulus values), excluding spectral information

Simulation software that supports this type of optical file is limited. For example, the Tracepro optical simulation software commonly used in the industry cannot use this type of optical file for color simulation. In addition, because each light does not contain spectral characteristics, if the lighting system contains reflective or dispersive materials or components, the dispersion effect of the light cannot be accurately simulated, and it is difficult to ensure the accuracy of the receiver.

Figure 4 Near-field model of light source containing only luminance, chrominance (tristimulus values) information opens in professional analysis software

c. Contains information on the distribution of luminance and chrominance, as well as spectral information

Most of the current mainstream simulation software supports this type of optical file, which can accurately simulate the chromaticity and luminosity distribution of the light source. Moreover, as shown in Figure 6, such a near-field model can obtain a spectral distribution of the source at any location in space.

Figure 5: Near-field model of the source containing spectral information opens the interface in professional analysis software.

Figure 6 Spectral distribution of a light source at a certain position in space

The b-type and c-type near-field models mentioned above can be simulated by Tracepro software to obtain the illuminance (intensity) distribution map shown in Figure 7. It has also been mentioned that the commonly used Tracepro optical simulation software in the industry cannot be used without spectrum. The near-field model document of the information is simulated in color, but when the spectral information is included, the Tracepro software above version 7.0 can accurately simulate the color of the light source. The difference between the two can be seen in Figure 8 a) and Figure 8 b ), the light source used in the figure is a warm white LED.

Figure 7 Illumination distribution map simulated by Tracepro software

Figure 8 a) TracePro simulation of true color map (without spectrum) Figure 8 b) TracePro simulation of true color map (with spectrum)

For the user, you can choose which type of near-field model to measure according to your needs. The SIG 400 of the Hong Kong University of Science and Technology Foshan LED Center contains spectrometer accessories, which can fully cover the three types of near-field tests mentioned above. How do you get the near field model?

The near-field model obtained by the near-field test cannot be directly applied to the optical simulation software. It is necessary to generate a ray set document containing the arbitrary number of rays corresponding to the optical simulation software through professional light source analysis software, and then import it into the corresponding optical simulation software for optical. Simulation, such as the commonly used simulation software LightTools, TracePro, ASAP, FRED, Zemax, LucidShape, Opticad, OSLO, SimuLux, SPEOS. Since the near-field model is a complete description of the light source, it can be used to generate and regenerate light sets at will (using less light in the initial design and using a lot of light in the final design), making the overall design process more efficient.

3. Test method for special light source

The SIG-400 light source near-field optical distribution test system (see Figure 9) of the Radiant Company of the Hong Kong University of Science and Technology Foshan LED Center is designed for small light sources such as LED lamp beads, COBs, modules, etc., and is equipped with the latest version of Radiant. ProSource 10.2.2, a professional light source analysis software, generates light sets for a variety of mainstream optical lighting design software. At the same time, for some special LED light sources, the center has been verified by independent research and can also achieve accurate measurement.

Figure 9 SIG-400 source near-field optical distribution test system Figure 10 ProSource source model analysis software

a. LED chip (bare crystal)

For optical measurement of LED chips, there are generally two types of measurement without package measurement and package measurement. If there is no package measurement, the direct contact with the chip is air, and the actual LED package application generally uses silicon dioxide to protect the chip and improve the light efficiency. The chip light will be affected by the light overflow angle, and the measured brightness and luminous flux will generally be There is a big difference in the brightness of the surface after encapsulation. Therefore, if the chip is directly in contact with air during the test, the near-field source model of the chip under use cannot be accurately obtained. If the package is measured, the shape and structure of the package will affect the light output of the LED itself, so the measurement results can only be valid for a specific package.

In response to the above problems, the Foshan LED Center of the Hong Kong University of Science and Technology provides a specific light source model structure for the near-field testing of LED chips (Figure 11). By simulating a simple package form, it is simultaneously measured by a near-field measuring device. The brightness distribution of different angles gives the near field distribution data of the light source. Since the relevant parameters of the glass sphere and the index matching liquid are known (the refractive index is very close to the refractive index of the common silica gel), the actual light-emitting condition of the surface of the LED chip can be obtained by the reverse tracing method, and then any packaging condition can be simulated in the optical software. Under the light effect.

Figure 11 bare metal near field test special fixture

b. High power light source (COB or module)

For low-power LEDs, it can be directly soldered to a common thin aluminum substrate to achieve heat dissipation. However, for high-power LEDs, the required heat sink is relatively large, even thicker. Because the distance between the sample and the colorimeter is very close when the near-field measurement is performed, if the heat sink of the sample is large or thick, the brightness meter cannot focus, and thus cannot Accurate measurement. To this end, the Foshan LED Center of the Hong Kong University of Science and Technology has independently developed a temperature control device that can solve the heat dissipation problem of high-power LEDs (up to 150W), thereby enabling near-field measurement of high-power LEDs. At the same time, the temperature control device can control the temperature range from 0 to 100 °C, which can meet the requirements of customers at the specified temperature.

c. Five-sided illuminated CSP source

At present, the popular CSP light source, especially the five-sided light source, has a small volume and some products have high power. If the welding wire is lit on a conventional aluminum substrate, the solder joints and the bonding wires may be partially blocked. Light, and it will cause some stray light. The Hong Kong University of Science and Technology Foshan LED Center has designed a special heat sink for this type of CSP light source, which perfectly solves the problem of light blocking and soldering of solder joints and wire bonds, and can perform near-field measurement more accurately.