A method for calibrating the luminous brightness of an AlGaInP quaternary LED chip production machine
By combining the calibration method of the luminous brightness mean and compensation value, the problem of low accuracy of the photoelectric parameter test of LED chip production machines is solved, and higher accuracy and stability are achieved to meet the market's demand for fine grading.
Patent Information
- Application Number
- CN202011184628.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-10-30
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2040-10-30
AI Technical Summary
The accuracy of the photoelectric parameter test of existing LED chip production machines is low, especially when calibration is performed with luminous brightness as the parameter. There are problems of inaccuracy and complex operation, which is difficult to meet the market's requirements for fine grading and stability.
Using a method combining the mean calibration of the luminous brightness and compensation value calibration, the average value of the standard sample is compared with the initial parameters of the production machine, and gradually corrected and compensated to improve the accuracy of calibration.
It significantly improves the test accuracy and consistency of LED chip production machines, enhances the stability of photoelectric parameters, and meets the market's requirements for fine grading and accuracy.
Smart Images

Figure CN114527362B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a luminous brightness calibration method for an AlGaInP quaternary LED chip production machine, belonging to the technical field of chip testing. Background Art
[0002] As the core component of semiconductor lighting, LED chips play an important role in the lighting industry. LED is known as the fourth generation of lighting sources or green light sources. It has the characteristics of energy saving, environmental protection, long life and small size. It is widely used in various indications, displays, decorations, backlight sources, general lighting and urban night scenes. According to different functions, it can be divided into five categories: information display, signal lights, automotive lamps, LCD screen backlights, and general lighting. LED chips have high luminous efficiency, a wide range of colors, and a long service life. They have been widely used in various fields such as large-screen display, landscape lighting, traffic lights, and car status displays. Therefore, the photoelectric performance parameters of LED chips directly determine the quality of their quality, and the quality of their photoelectric performance is determined by testing. How to ensure the accuracy and stability of the photoelectric parameters of LED chips has become a key criterion for judging the quality of LED chips.
[0003] With the rapid development of market demand and packaging technology, the LED chip market has put forward higher requirements for the optoelectronic parameters of the chip. Therefore, how to ensure the accuracy and stability of the optoelectronic parameter testing of LED chips is becoming more and more important. The different sizes of LED chips, electrode shapes, substrate types, wafer cutting expansion spacing, test machine light collection architecture and other process factors have an impact on the optoelectronic parameters, which will affect the mutual calibration between various machines. Therefore, how to ensure the consistency and accuracy of the tests of different equipment is the key to the test. With the advancement of technology and the lean needs of the market, the smaller and more refined the range of LED chip grading, the higher the requirements for the stability and accuracy of the test data. The various calibration methods currently used in the industry to calibrate the accuracy of the machine are mostly through wavelength multi-band calibration compensation to correct the data within a certain error range.
[0004] At present, wavelength multi-band calibration (using wavelength as data processing) is mostly used to compensate and adjust the test machine. For example, Chinese patent document CN105388439A provides a multi-band calibration method for LED chip testing, including the steps of: making a standard sample; using a standard machine to test the standard sample to obtain the wavelength of each band of the standard sample; using a standard machine to determine the starting point and point measurement direction of the standard sample; using a production machine to test the standard sample to obtain the wavelength of each band of the standard sample; data comparison adopts point-to-point corresponding sub-band comparison: the wavelength of the standard sample tested by the standard machine and the difference are made into a scatter plot; each band is corrected according to the scatter plot; and the production machine and the standard machine are calibrated to be consistent in each band according to the correction method. However, multi-band calibration is more convenient for yellow-green light that is more sensitive to wavelength. For the four-element red-yellow light with a wider wavelength of AlGaInP, the multi-band calibration steps are more complicated and the operation is not simple. Calibration can be performed using luminous brightness as a parameter.
[0005] At present, there are also patent documents reporting on the method of calibration using luminous brightness as a parameter. For example: Chinese patent CN104849683A discloses a method for calibrating LED chip measurement equipment, which first selects the grains required for calibration as standard grains; the test standard machine measures the standard grains to obtain the initial value A; the IS standard measurement equipment obtains the standard value B; the brightness column of the standard value B is used to make a scatter plot for the brightness column of the initial value A, and the fitting function is obtained to obtain the calibration value of the brightness. The fitting curve can use a single fit or a double fit according to the needs. The single fit obtains the fitting function y=ax+b, where a and b are the calibration values of the brightness. However, the calibration data processing of the above method directly uses the standard value and the test initial value to compare the scatter plot. With the repeated calibration of the test machine and the setting of the test parameters, it will cause differences in the initial values of each test, resulting in inaccuracy in the brightness calibration value generated by the calibration, resulting in errors in the test.
[0006] At present, the method of directly using the luminous brightness as a parameter to calibrate the machine has low accuracy. Therefore, it is urgent to develop a luminous brightness calibration method for LED chip production machines with high accuracy and simple operation. Summary of the invention
[0007] In view of the deficiencies in the prior art, the present invention provides a luminous brightness calibration method for an AlGaInP quaternary LED chip production machine. The method of the present invention performs calibration based on luminous brightness, combines a luminous brightness mean calibration method with a calibration method for increasing a compensation value, firstly corrects the production machine through basic luminous brightness mean calibration, and then, based on the existing calibration method for increasing a compensation value, further converts the luminous brightness standard value into a luminous brightness test value under the initial parameter setting of the production machine for comparison and calibration, rather than directly comparing and calibrating the luminous brightness standard value with the luminous brightness test value under the existing setting of the production machine, thereby thoroughly improving the accuracy of the test calibration from the calibration method.
[0008] Terminology Note:
[0009] Gain value: a corresponding to y=ax and y=ax+b is the mean correction value of brightness calibration, where x is the average luminance of the standard sample tested on the production machine, and y is the average luminance of the standard sample tested on the standard machine.
[0010] Offset value: b corresponding to y=ax+b is the compensation value of brightness calibration, where x and y are the same as described above.
[0011] The technical solution of the present invention is as follows:
[0012] A method for calibrating the luminous brightness of an AlGaInP quaternary LED chip production machine comprises the following steps:
[0013] (1) Making a standard sample, testing the standard sample using a standard machine to obtain the luminous brightness A of each grain in the standard sample, and calculating the average value of the luminous brightness A; using the standard machine to determine the starting point and point measurement direction of the standard sample test;
[0014] (2) Use standard samples to calibrate the average brightness of the production machine
[0015] The standard sample is tested by the production machine to obtain the luminous brightness B of each grain in the standard sample, and the abnormal points in the test are statistically screened and removed to calculate the average value of the luminous brightness B; the average value of the luminous brightness B of the standard sample is compared with the average value of the luminous brightness A. If the difference rate between the average value of the luminous brightness B of the standard sample and the average value of the luminous brightness A is ≤2%, the production machine does not need to be calibrated; if the difference rate between the average value of the luminous brightness B of the standard sample and the average value of the luminous brightness A is greater than 2%, the Gain value of the production machine needs to be calibrated according to the following process:
[0016] During the calibration process, y = the average value of the luminous brightness A of the standard sample, x = the average value of the luminous brightness B of the standard sample, a = the Gain value before the production machine is calibrated. If x>y, the Gain value after calibration (a) = [1-(xy) / y]×a; if x<y, the Gain value after calibration (a) = [1+|(xy) / y|]×a;
[0017] (3) Use standard samples to further correct the average brightness of the production machine and compensate for deviations
[0018] After setting the Gain value of the production machine to the Gain value (a) calibrated in step (2), the production machine is used to test the standard sample again to obtain the luminous brightness C of each grain in the standard sample, statistically screen and remove abnormal points in the test, and compare the luminous brightness C and luminous brightness A of each grain in the standard sample one by one, and calculate the difference rate of each point; if the average value of the difference rate between the luminous brightness C and the luminous brightness A is ≤1%, there is no need to further calibrate the production machine; if the average value of the difference rate between the luminous brightness C and the luminous brightness A is >1%, proceed to the next step to further calibrate the production machine;
[0019] (4) Set the Gain value of the luminous brightness (LOP1) calibration point of the production machine to 1 and the Offset value to 0, and use the production machine to test the standard sample again. Statistically screen and remove abnormal points in the test to obtain the luminous brightness D of each grain in the standard sample. Make a scatter plot with the luminous brightness D data of each grain in the standard sample as the x-axis and the luminous brightness A data as the y-axis, and perform linear fitting to obtain the fitting line and the formula y=ax+b. In the formula, a is the Gain value after the final calibration, and b is the Offset value. The calibration is completed.
[0020] Preferably, according to the present invention, the standard sample described in step (1) is a chip with uniform photoelectric parameters, and the standard sample meets the following conditions: the luminous brightness fluctuation range is ≤320mcd, the voltage fluctuation range is ≤0.02V, and the wavelength fluctuation range is ≤0.5nm; the selected chip is tested by a standard machine, and then the partial area that meets the above conditions is cut out to obtain a standard sample.
[0021] Preferably according to the present invention, the number of grains in the standard sample in step (1) is 200-500.
[0022] Preferably, according to the present invention, in step (2), the starting point and point measurement direction of the standard sample tested by the production machine are consistent with the starting point and point measurement direction in step (1), so as to facilitate point-to-point comparison of data.
[0023] Preferably, according to the present invention, the calculation method of the difference rate in step (2) is: (|average value of luminous brightness B-average value of luminous brightness A| / average value of luminous brightness A)×100%.
[0024] Preferably, according to the present invention, in step (3), the starting point and point measurement direction of the standard sample tested by the production machine are consistent with the starting point and point measurement direction in step (1), so as to facilitate point-to-point comparison of data.
[0025] Preferably, according to the present invention, the calculation method of the difference rate in step (3) is: (|luminous brightness C-luminous brightness A| / luminous brightness A)×100%.
[0026] Preferably, according to the present invention, in step (4), the starting point and point measurement direction of the standard sample tested by the production machine are consistent with the starting point and point measurement direction in step (1), so as to facilitate point-to-point comparison of data.
[0027] Preferably according to the present invention, the abnormal point described in steps (2), (3) and (4) refers to a point where the voltage is greater than 3V, the luminous brightness is less than 10mcd, the wavelength is greater than 630nm or less than 615nm, and the reverse leakage current is greater than 1μA.
[0028] The Gain value in the present invention refers to the Gain value at the calibration point of the light luminance (LOP1) of the production machine.
[0029] Everything that is not described in detail in the present invention is prior art in the art.
[0030] The technical features and beneficial effects of the present invention are as follows:
[0031] 1. The present invention first uses the average value of luminous brightness to calibrate the production machine, corrects the deviation that has occurred in the machine production operation, plays a basic calibration role, and lays a good foundation for further testing and calibration.
[0032] 2. After the present invention calibrates according to the mean luminous brightness, it further restores the calibration value of the production machine to the original setting. The original setting is to directly restore the Gain value and Offset value that the machine has accumulated to the original setting, that is, the Gain value is set to 1, and the Offset value is set to 0. After eliminating the accumulated error, recalibration is performed, and the obtained test data is then compared with the standard data point by point. The machine deviation is further calibrated from the original value, and the compensation value is added to further correct and compensate for the deviation of the mean luminous brightness, thereby significantly improving the accuracy of the calibration.
[0033] 3. After calibrating the production machine using the calibration method described in the present invention, the test accuracy and consistency are significantly improved, the accuracy and stability of the LED chip photoelectric parameter test are improved, and the judgment basis for market customer needs is improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 Schematic diagram of the arrangement of grains in the standard sample in Example 1.
[0035] Figure 2 This is a fitting curve diagram of the test data after the production machine in Example 1 is restored to its original settings. DETAILED DESCRIPTION
[0036] The present invention will be described in detail below with reference to specific embodiments and the accompanying drawings, but is not limited thereto.
[0037] Anything not described in detail in the present invention is conventional technology in the art.
[0038] Meanwhile, the instruments and equipment used in the following embodiments, unless otherwise specified, are commonly used instruments and equipment in the art and can be purchased commercially.
[0039] Example 1
[0040] A method for calibrating the luminous brightness of an AlGaInP quaternary LED chip production machine comprises the following steps:
[0041] (1) Making a standard sample: Select a chip with uniform photoelectric parameters, test the selected chip using a standard machine, and then cut out a part of the area that meets the following conditions: luminous brightness fluctuation range ≤ 320mcd, voltage fluctuation range ≤ 0.02V, wavelength fluctuation range ≤ 0.5nm to obtain a standard sample. The number of grains in the standard sample is 200, and the arrangement of the grains is shown in the schematic diagram. Figure 1 As shown;
[0042] The standard sample was tested using a standard machine to obtain the luminous brightness A of each grain in the standard sample, and the average value of the luminous brightness A was calculated to be 5369mcd; the starting point and point measurement direction of the standard sample test were determined using the standard machine.
[0043] (2) Use standard samples to calibrate the average brightness of the production machine
[0044] According to the starting point and point measurement direction of the standard sample test determined in step (1), the standard sample is tested using the production machine to obtain the luminous brightness B of each grain in the standard sample, and the abnormal points in the test (voltage greater than 3V, luminous brightness less than 10mcd, wavelength greater than 630nm or less than 615nm, reverse leakage current greater than 1μA) are statistically screened and removed, and the average value of the luminous brightness B is calculated to be 5214mcd; the average value of the luminous brightness B of the standard sample is compared with the average value of the luminous brightness A, and the difference rate between the average value of the luminous brightness B of the standard sample and the average value of the luminous brightness A is (|5214-5369| / 5369)×100%=2.89%>2%, then the Gain value of the production machine is calibrated according to the following process:
[0045] During the calibration process, y = the average value of the luminous brightness A of the standard sample, 5369; x = the average value of the luminous brightness B of the standard sample, 5214; a = the Gain value before calibration of the production machine, 7.38; where y = 5369, x = 5214, because x < y, the Gain value after calibration (a) = [1 + | (5214 - 5369) | / 5369 |] × 7.38 = 7.59.
[0046] (3) Use standard samples to further correct the average brightness of the production machine and compensate for deviations
[0047] After setting the Gain value of the production machine to the Gain value (a) 7.59 calibrated in step (2), the standard sample is tested again using the production machine according to the starting point and point measurement direction of the standard sample test determined in step (1) to obtain the luminous brightness C of each grain in the standard sample, statistically screening and removing abnormal points in the test (voltage greater than 3V, luminous brightness less than 10mcd, wavelength greater than 630nm or less than 615nm, reverse leakage current greater than 1μA), and the luminous brightness C and luminous brightness A of each grain in the standard sample are compared point by point; after the one-to-one comparison, the difference rate of each point is calculated according to the formula of difference rate = (|luminous brightness C-luminous brightness A| / luminous brightness A) × 100%, and the data comparison is shown in Table 1; after calculation, the average value of the difference rate between luminous brightness C and luminous brightness A is 1.40%>1%, then step (4) is performed to further calibrate the production machine;
[0048] Table 1 Point-to-point comparison data of luminous brightness C and luminous brightness A
[0049]
[0050]
[0051]
[0052]
[0053]
[0054] (4) Set the Gain value of the calibration point of the luminous brightness (LOP1) of the production machine to 1, and the Offset value to 0. According to the starting point and point measurement direction of the standard sample test determined in step (1), use the production machine to test the standard sample again, statistically screen and remove abnormal points in the test (voltage greater than 3V, luminous brightness less than 10mcd, wavelength greater than 630nm or less than 615nm, reverse leakage current greater than 1μA), and obtain the luminous brightness D of each grain in the standard sample. Use the luminous brightness D data of each grain in the standard sample as the x-axis and the luminous brightness A data as the y-axis to make a scatter plot, perform linear fitting, and obtain the fitting straight line and the formula y=7.665x+3220, as shown in Figure 2 As shown, a in the formula is the Gain value after final calibration, and b is the Offset value. Fill a, i.e. 7.665, in the final formula into the Gain value of production machine LOP1, and fill b, i.e. 3220, into the Offset value of LOP1 to complete the calibration.
Claims
1. A method for calibrating the luminous brightness of an AlGaInP quaternary LED chip production machine, comprising the following steps: (1) Prepare a standard sample, use a standard machine to test the standard sample to obtain the luminous brightness A of each grain in the standard sample, and calculate the average value of the luminous brightness A; use the standard machine to determine the starting point and point measurement direction of the standard sample test; (2) Use standard samples to calibrate the average luminous brightness of the production machine The standard sample is tested by the production machine to obtain the luminous brightness B of each grain in the standard sample, and the abnormal points in the test are statistically screened and removed to calculate the average value of the luminous brightness B; the average value of the luminous brightness B of the standard sample is compared with the average value of the luminous brightness A. If the difference rate between the average value of the luminous brightness B of the standard sample and the average value of the luminous brightness A is ≤2%, the production machine does not need to be calibrated; if the difference rate between the average value of the luminous brightness B of the standard sample and the average value of the luminous brightness A is greater than 2%, the Gain value of the production machine needs to be calibrated according to the following process: During the calibration process, y = the average value of the luminance A of the standard sample, x = the average value of the luminance B of the standard sample, a = the Gain value before calibration of the production machine. If x>y, the Gain value after calibration (a) = [1-(xy) / y]×a; if x<y, the Gain value after calibration (a) = [1+|(xy) / y|]×a; (3) Use standard samples to further correct the average brightness of the production machine and compensate for deviations After setting the Gain value of the production machine to the Gain value (a) calibrated in step (2), the production machine is used to test the standard sample again to obtain the luminous brightness C of each grain in the standard sample, statistically screen and remove abnormal points in the test, and compare the luminous brightness C and luminous brightness A of each grain in the standard sample one by one, and calculate the difference rate of each point; if the average value of the difference rate between the luminous brightness C and the luminous brightness A is ≤1%, there is no need to further calibrate the production machine; if the average value of the difference rate between the luminous brightness C and the luminous brightness A is >1%, proceed to the next step and further calibrate the production machine; (4) Set the Gain value at the luminous brightness calibration point of the production machine to 1 and the Offset value to 0, and use the production machine to test the standard sample again. Statistically screen and remove abnormal points in the test to obtain the luminous brightness D of each grain in the standard sample. Make a scatter plot with the luminous brightness D data of each grain in the standard sample as the x-axis and the luminous brightness A data as the y-axis, and perform linear fitting to obtain the fitting line and the formula y=ax+b. In the formula, a is the Gain value after the final calibration, and b is the Offset value. The calibration is completed.
2. The luminous brightness calibration method of the AlGaInP quaternary LED chip production machine according to claim 1, characterized in that: The standard sample described in step (1) meets the following conditions: the luminous brightness fluctuation range is ≤320mcd, the voltage fluctuation range is ≤0.02V, and the wavelength fluctuation range is ≤0.5nm.
3. The luminous brightness calibration method of the AlGaInP quaternary LED chip production machine according to claim 1, characterized in that: The number of grains in the standard sample described in step (1) is 200-500.
4. The luminous brightness calibration method of the AlGaInP quaternary LED chip production machine according to claim 1, characterized in that: In step (2), the starting point and point measurement direction of the standard sample tested by the production machine are consistent with the starting point and point measurement direction in step (1).
5. The luminous brightness calibration method of the AlGaInP quaternary LED chip production machine according to claim 1, characterized in that: The calculation method of the difference rate in step (2) is: (|average value of luminous brightness B - average value of luminous brightness A| / average value of luminous brightness A)×100%.
6. The luminous brightness calibration method of the AlGaInP quaternary LED chip production machine according to claim 1, characterized in that: In step (3), the starting point and point measurement direction of the standard sample tested by the production machine are consistent with the starting point and point measurement direction in step (1).
7. The luminous brightness calibration method of the AlGaInP quaternary LED chip production machine according to claim 1, characterized in that: The calculation method of the difference rate in step (3) is: (|luminous brightness C-luminous brightness A| / luminous brightness A)×100%.
8. The luminous brightness calibration method of the AlGaInP quaternary LED chip production machine according to claim 1, characterized in that: In step (4), the starting point and point measurement direction of the standard sample tested by the production machine are consistent with the starting point and point measurement direction in step (1).
9. The luminous brightness calibration method of the AlGaInP quaternary LED chip production machine according to claim 1, characterized in that: The abnormal point described in steps (2), (3) and (4) refers to a point where the voltage is greater than 3V, the luminance is less than 10mcd, the wavelength is greater than 630nm or less than 615nm, and the reverse leakage current is greater than 1μA.
