CN109060164B - Light-emitting device temperature distribution measurement device and measurement method based on micro-hyperspectral - Google Patents

Abstract

A light-emitting device temperature distribution measurement device and measurement method based on micro-hyperspectrum belongs to the field of testing the temperature distribution of light-emitting devices, including a temperature control platform, a driving power supply, a temperature control power supply, a microscope, a hyperspectral instrument and a computer; the initial temperature of a sample is set, and the The driving power supply selects the pulse signal to drive the sample; adjust the temperature control power supply, change the temperature of the temperature control platform, and the hyperspectral instrument collects the hyperspectral data of the sample at the corresponding temperature point; the computer calculates the two-dimensional temperature sensitivity coefficient matrix according to the hyperspectral data; To the constant voltage or constant current mode, use the hyperspectral instrument to collect the hyperspectral data of the sample; the computer uses the hyperspectral data under the constant voltage or constant current mode to calculate the centroid wavelength, and combine the two-dimensional temperature sensitivity coefficient matrix to obtain the two-dimensional temperature distribution of the sample surface ; The spectral image of each pixel on the surface of the light-emitting device can be obtained, so as to accurately obtain the two-dimensional temperature distribution map of the surface of the light-emitting device, and intuitively reflect the trend of its surface temperature change.

Description

Light-emitting device temperature distribution measurement device and measurement method based on micro-hyperspectral

technical field

The invention belongs to the field of testing the temperature distribution of light-emitting devices, in particular to a measuring device and method for measuring the temperature distribution of light-emitting devices based on microscopic hyperspectral.

Background technique

In the era of energy economy in the 21st century, light-emitting devices such as semiconductor light-emitting diodes (LEDs) are widely used in many fields such as daily life, electronic technology, and military affairs because of their small size, long life, and fast response. LED converts electrical energy into light energy, but since 60% to 70% of electrical energy will be converted into heat energy, if a large amount of heat cannot be dissipated, the junction temperature will be too high, which will seriously affect the service life of LED. Therefore, the research on LED junction temperature has far-reaching significance.

At present, the measurement methods for the junction temperature of light-emitting devices are indirect tests using the relationship between a certain characteristic of the device itself and the temperature, mainly including forward voltage drop method, spectral blue-white ratio method, spectral peak wavelength method, spectral centroid wavelength and half-height Broad and infrared thermal imaging, etc.

The forward voltage drop method is currently commonly used to measure the junction temperature. It uses the linear relationship between the voltage at both ends of the device and the temperature to test the junction temperature, but its test accuracy is limited by the switching speed from heating a large current to testing a small current. , and for the finished lamps after packaging, due to many limitations such as the material of the lamp shell, it is generally difficult to accurately measure the voltage drop on each LED pin ([1] Ryu H Y, Ha K H, Chae J H, et al.Measurement of junction temperature in GaN-based laser diodes using voltage-temperature characteristics[J].Applied Physics Letters,2005,87(9):297.).

The main principle of the spectral blue-white ratio method is that as the junction temperature rises, the luminescence of the chip and the photoluminescence of the phosphor decrease at the same time, but the luminescence of the phosphor decreases more rapidly, so that the proportion of blue-white light in the white light spectrum changes. . The disadvantage of this method is that it is not suitable for the measurement of monochromatic LED junction temperature ([2] Tetsushi Tamura, Tatsumi Setomoto, TsunemasaTaguchi. Illumination characteristics of lighting array using 10candela-classwhite LEDs under AC 100V operation [J]. Journal of Luminescence, 2000, 87.).

The spectral peak wavelength method measures the LED junction temperature through the drift relationship of the peak wavelength with the junction temperature. The disadvantage of this method is that the change of the peak wavelength of the spectrum is small, resulting in a large actual measurement error, which is greatly affected by the experimental instrument, and the variation of the peak wavelength of different LEDs with temperature is inconsistent ([3] Y.Xi, J.Q.Xi, T.Gessmann, J.M. Shah, J.K.Kim, E.F.Schubert, A.J.Fischer, M.H.Crawford, K.H.A.Bogart, and A.A.Allerman, "Junction and carrier temperature measurements in deep-ultraviolet light-emitting diodes using three different methods," Appl. Phys. Lett., vol.86, no.3, pp.031907, Jan.2005.).

The spectral centroid wavelength and FWHM method is based on the existence of a mathematical model relationship between optical parameters (centroid wavelength and FWHM) and junction temperature and DC drive current. This method is to reduce the temperature of the heat sink while increasing the DC drive current, so as to ensure that the overall junction temperature is constant, and correct the functional relationship among the optical parameters, current and junction temperature, so as to obtain the expression of the junction temperature Formula ([4] Y.Lin, Y.L.Gao, Y.J.Lu, L.H.Zhu, Y.Zhang and Z.Chen, "Study of temperature sensitive optical parameters and junction temperature determination of light-emitting diodes," Appl. Phys. Lett., vol .100, pp.202108, May.2012.).

The above-mentioned methods are all aimed at testing the average temperature of the light-emitting device, and only one average temperature can be obtained to represent the temperature of the light-emitting device. However, the surface temperature of the light-emitting device is distributed in two dimensions, and the temperature distribution at each position is not completely consistent. The infrared thermal imaging method measures the two-dimensional temperature distribution of the surface through the infrared radiation on the surface of the object. The disadvantage is that it is easily affected by the package structure of the device, and it is difficult to detect the surface of the packaged LED chip ([5]Tamdogan E, Pavlidis G, Graham S, et al. .A Comparative Study on the Junction Temperature Measurements of LEDs With Raman Spectroscopy,Microinfrared(IR)Imaging,and Forward Voltage Methods[J].IEEE Transactions on Components Packaging&Manufacturing Technology,2018,PP(99):1-9.).

Hyperspectral technology is an emerging comprehensive detection technology in recent years. It organically composes image information and spectral data into a three-dimensional data cube that integrates maps and spectra, and integrates imaging technology, electronic technology, information technology, computer technology and other technologies. Compared with traditional means, it has a more powerful data acquisition function. It can combine spectral information and spatial information, and has been widely used in remote sensing, agronomy, medical treatment, criminal investigation, identification and other fields.

Contents of the invention

The purpose of the present invention is to solve the above-mentioned problems in the prior art, and to provide a light-emitting device temperature distribution measurement device and measurement method based on microscopic hyperspectral, which can accurately obtain the spectral image of each pixel on the surface of the light-emitting device, thereby accurately obtaining the temperature distribution of the light-emitting device. The surface two-dimensional temperature distribution map intuitively reflects the surface temperature change trend.

To achieve the above object, the present invention adopts the following technical solutions:

A light-emitting device temperature distribution measurement device based on microscopic hyperspectral, including a temperature control platform, a driving power supply, a temperature control power supply, a microscope, a hyperspectral instrument and a computer, and the temperature control platform is used to fix samples;

The driving power is connected to the sample, and the driving power is used to drive the sample;

The temperature control power supply is connected to the temperature control platform, and the temperature control power supply is used to control the temperature of the sample;

The microscope is set above the sample, and the microscope is used to transmit the light emitted by the sample;

The hyperspectral instrument is set above the microscope, and the hyperspectral instrument is used to collect the light transmitted by the microscope and obtain hyperspectral data;

The computer is connected to the hyperspectral instrument, and the computer is used to calculate the received hyperspectral data to obtain the two-dimensional temperature distribution of the sample surface.

The driving power of the present invention includes pulse voltage mode, pulse current mode, constant voltage mode and constant current mode.

A method for measuring temperature distribution of a light-emitting device based on microscopic hyperspectral, comprising the following steps:

Step 1. Fix the sample on the temperature control platform, set the initial temperature T ₀ of the sample, and control the temperature of the sample at T ₀ by using the temperature control power supply;

Step 2. Adjust the driving power to pulse voltage or pulse current mode, and select a short pulse signal to drive the sample;

Step 3. Adjust the temperature control power supply, and change the temperature of the temperature control platform to T ₀ , T ₁ , T ₂ . ;

Step 4, the computer fits the relationship between the centroid wavelength of the sample and the sensitivity coefficient of the surface temperature according to the hyperspectral data, and then obtains a two-dimensional temperature sensitivity coefficient matrix;

Step 5. Adjust the driving power to constant voltage mode or constant current mode. The voltage or current corresponds to the high-level voltage or current under the pulse. The light emitted by the sample enters the hyperspectral instrument through the microscope, and the hyperspectral instrument is used to collect constant voltage or constant current. The spectral data of the sample in the mode;

Step 6. The computer uses the hyperspectral data in constant voltage mode or constant current mode to calculate the centroid wavelength, and combines the two-dimensional temperature sensitivity coefficient matrix obtained in step 4 to obtain the two-dimensional temperature distribution on the surface of the sample.

Compared with the prior art, the beneficial effects obtained by the technical solution of the present invention are:

1. The present invention uses microscopic hyperspectral to obtain the principle of linear relationship between the spectral centroid wavelength of each pixel point of the light-emitting device and the temperature change to test the two-dimensional temperature distribution of the light-emitting device. Compared with other junction temperature testing methods, the spatial resolution of microscopic hyperspectral technology With high efficiency and spectral resolution, the spectral image of each pixel on the surface of the light-emitting device can be accurately obtained, so that the two-dimensional temperature distribution map of the surface of the light-emitting device can be accurately obtained, and the surface temperature change trend can be intuitively reflected.

2. The invention has a wide range of applications and is not affected by the packaging of the light-emitting device. No matter whether the light-emitting device itself is packaged or whether it is covered with other materials, as long as the light can pass through the package, the two-dimensional temperature distribution on the surface of the light-emitting chip can be obtained.

Description of drawings

Fig. 1 is a schematic diagram of the workflow of the present invention;

Fig. 2 is a graph of the linear fitting relationship between the centroid wavelength and the temperature of the blue LED;

Figure 3 is a two-dimensional temperature distribution diagram of the surface of the blue LED calculated under constant voltage mode at 25°C;

Fig. 4 is a longitudinal temperature change trend diagram at position A of the blue LED in Fig. 3;

Fig. 5 is a trend diagram of lateral temperature change at position B of the blue LED in Fig. 3;

Fig. 6 is a statistical diagram of the two-dimensional temperature points on the surface of the blue LED calculated under the constant voltage mode at 25°C.

Reference signs: 1-drive power supply, 2-temperature-controlled power supply, 3-temperature-controlled platform, 4-sample, 5-microscope, 6-hyperspectrometer, 7-computer.

Detailed ways

In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention clearer and clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

As shown in Figure 1, the present invention comprises driving power supply 1, temperature control power supply 2, temperature control table 3, microscope 5, hyperspectrometer 6 and computer 7; Wherein:

The temperature control platform 3 is used to fix the sample 4;

The driving power supply 1 is connected to the sample 4, and the driving power supply 1 is used to drive the sample 4; the driving power supply 1 includes a pulse voltage mode, a pulse current mode, a constant voltage mode and a constant current mode;

The temperature control power supply 2 is connected to the temperature control platform 3, and the temperature control power supply 2 is used to control the temperature of the sample 4;

The microscope 5 is arranged above the sample 4, and the microscope 5 is used to pass through the light emitted by the sample 4;

The hyperspectral instrument 6 is arranged above the microscope 5, and the hyperspectral instrument 6 is used to collect the light transmitted by the microscope 5 and obtain hyperspectral data;

The computer 7 is connected to the hyperspectral instrument 6, and the computer 7 is used to calculate the received hyperspectral data to obtain the two-dimensional temperature distribution on the surface of the sample 4.

The sample described in the present invention is a light emitting device.

The invention combines hyperspectral technology and microscopic technology to obtain high-resolution spectral information of each pixel on the surface of a light-emitting device, utilizes the linear relationship between centroid wavelength and temperature change, and uses voltage or current pulse signals to correct temperature.

The centroid wavelength is the geometrically symmetric wavelength of the spectral distribution, defined as follows:

Where: F(λ) is the spectral distribution of the light-emitting device; λ is the wavelength, λ ₁ and λ ₂ are the upper and lower limit wavelengths of the spectral distribution, and the range of visible light is generally 380-780nm.

The centroid wavelength-temperature sensitivity coefficient K _ij of any pixel point is:

Among them: T ₀ and T are the initial temperature of the temperature control platform and another control temperature respectively; i=1...x, j=1...y, x, y are the imaging dimension pixels of the hyperspectral instrument, Δλ _cxy is the xy pixel point at T temperature The difference from the centroid wavelength at T ₀ temperature.

The two-dimensional temperature sensitivity coefficient matrix of all pixels on the surface is:

The difference matrix between the centroid wavelengths of all pixels on the surface of the device at the temperature to be measured and the initial temperature is:

Then, the two-dimensional temperature distribution matrix of all pixels on the surface of the light emitting device is:

In this embodiment, the blue light LED is used as the sample to be tested, and the measurement method of the present invention is adopted, which specifically includes the following steps:

Step 1. Fix the blue LED on the temperature control table 3, and use the temperature control power supply 2 to control the temperature of the heat sink of the sample 4 at 25°C, and the blue LED is not powered.

Step 2. Adjust the driving power supply 1 so that the signal is: the pulse period is 2ms, the high level is 3V, the low level is 0V, and the duty ratio is 3%. , can better avoid the influence of self-heating effect).

Step 3, adjust the temperature control power supply 2, sequentially control the temperature of the heat sink of the sample 4 to 25°C, 35°C, 45°C, 55°C, 65°C, 75°C, the light emitted by the sample 4 enters the hyperspectral instrument 6 through the microscope 5, and the hyperspectral instrument 6 collect blue light LED hyperspectral data to computer 7.

Step 4, the computer 7 calculates the centroid wavelength of each pixel point on the chip surface from the hyperspectral data, and fits the centroid wavelength-temperature sensitivity relational expression of each pixel point according to formula (1), as shown in Figure 2. The temperature linear fitting relationship curve, and then obtain the two-dimensional temperature sensitivity coefficient matrix K of all pixels shown in the formula (2).

Step 5. Adjust the driving power supply 1 again, and power on the blue LED in the constant voltage mode corresponding to the pulse high level of 3V.

Step 6. Repeat step 3 to obtain the hyperspectral data of blue LEDs at heat sink temperatures of 25°C, 35°C, 45°C, 55°C, 65°C, and 75°C under a constant voltage of 3V.

Step 7. Extract the centroid wavelength from the obtained hyperspectral data in the constant pressure mode. According to the two-dimensional temperature sensitivity coefficient matrix K obtained in step 4 and the formulas (3) and (4), the constant pressure and different heat sinks are obtained. The two-dimensional temperature distribution diagram of the surface of the blue LED at different temperatures, as shown in Figure 3, is the two-dimensional temperature distribution diagram of the surface of the blue LED calculated under the constant voltage mode of 25°C. Figure 4 and Figure 5 are graphs showing the longitudinal and lateral temperature change trends of blue LEDs. Due to the large number of temperature points on the two-dimensional surface, in order to compare with the results of thermocouples and verify the reliability of the present invention, the median temperature of the group with the highest frequency of occurrence in the group whose temperature interval is 0.1 ° C, that is, the mode temperature, mode The temperature is a representative surface temperature under this condition. Figure 6 is a statistical diagram of the two-dimensional temperature points on the surface of the blue LED calculated under the constant voltage mode of 25°C.

Step 8. Use a thermocouple to measure the surface temperature of the blue LED when the heat sink temperature is 25°C, 35°C, 45°C, 55°C, 65°C, and 75°C under a constant voltage of 3V, which is the same as that calculated in step 7. Compared with the mode temperature, the average error is less than 0.7°C, which proves the reliability of the method.

Table 1 (℃)

heat sink temperature 25 35 45 55 65 75 average error The temperature measured by this method 34.0 46.6 56.8 66.6 77.0 88.1 / Thermocouples measure temperature 35.1 45.6 54.9 66.1 75.3 88.0 / single random error -1.1 1.0 1.9 0.5 1.7 0.1 0.7

Table 1 shows the comparison results between the representative surface temperature (mode temperature) obtained by the microscopic hyperspectral technique and the surface temperature measured by the thermocouple of the blue LED at different temperatures.

Claims (1)

1. A light-emitting device temperature distribution measuring method based on microscopic hyperspectrum is characterized in that: the measuring device adopted by the measuring method comprises a temperature control table, a driving power supply, a temperature control power supply, a microscope, a hyperspectral meter and a computer;

The temperature control table is used for fixing a sample;

The driving power supply is connected with the sample and is used for driving the sample; the driving power supply comprises a pulse voltage mode, a pulse current mode, a constant voltage mode and a constant current mode;

the temperature control power supply is connected with the temperature control table and is used for controlling the temperature of the sample;

The microscope is arranged above the sample and is used for transmitting light emitted by the sample;

The hyperspectral meter is arranged above the microscope and used for collecting light transmitted by the microscope and obtaining hyperspectral data;

the computer is connected with the hyperspectral spectrometer and used for calculating the received hyperspectral data to obtain two-dimensional temperature distribution of the surface of the sample;

the measuring method comprises the following steps:

Step 1, fixing a sample on a temperature control table, setting the initial temperature T0 of the sample, and controlling the temperature of the sample to be T0 by using a temperature control power supply;

step 2, adjusting a driving power supply to a pulse voltage or pulse current mode, and selecting a short pulse signal to drive a sample;

step 3, adjusting a temperature control power supply, sequentially changing the temperature of a temperature control table to be T0, T1 and T2 … …, enabling light emitted by a sample to enter a hyperspectral meter through a microscope, and acquiring hyperspectral data of the sample at a corresponding temperature point by the hyperspectral meter;

Step 4, fitting the sensitivity coefficient relation between the centroid wavelength and the surface temperature of the sample by the computer according to the hyperspectral data to obtain a two-dimensional temperature sensitivity coefficient matrix;

step 5, adjusting the driving power supply to a constant voltage mode or a constant current mode, wherein the voltage or current corresponds to high level voltage or current under the pulse, light emitted by the sample enters the hyperspectral meter through the microscope, and the hyperspectral meter is used for collecting spectral data of the sample under the constant voltage or constant current mode;

and 6, calculating the centroid wavelength by using hyperspectral data in a constant voltage mode or a constant current mode by using a computer, and obtaining the two-dimensional temperature distribution of the surface of the sample by combining the two-dimensional temperature sensitivity coefficient matrix obtained in the step 4.