Mini LED chip sorting method and system

Abstract

The invention discloses a Mini LED chip sorting method and system, and relates to the technical field of chip sorting, and the method comprises the steps: carrying out the spot test of crystal grains in a wafer, so as to generate test data; filling neutral position data formed by spot test according to the test data, so that all crystal grains in the wafer correspond to the test data; carrying out cutting treatment on the wafer so as to cut crystal grains in the wafer into chips; carrying out expansion processing on the wafer so as to enlarge the interval of the chips; performing AOI detection on the wafer to generate detection data; sorting the chips in the wafer according to the test data and the detection data so as to classify the chips to corresponding grades; and performing visual inspection processing on the wafer to identify the appearance defects of the sorted chips. By adopting the method, the abnormal crystal grains can be effectively clamped and controlled, the abnormal crystal grains are clamped in a factory, the abnormal PPM value of the crystal grains is reduced, customer complaints are reduced, and the product quality is improved.

Description

Technical Field

[0001] The present invention relates to the field of chip sorting technology, and in particular to a Mini LED chip sorting method and system. Background Art

[0002] As the production capacity requirements become higher and higher and Mini LED products are mass-produced, the LED chip manufacturing industry uses sampling to test wafers. Therefore, there is no test data for each grain in the wafer, only the sampling data of some grains. Therefore, in the sorting process of the back-end, it is impossible to 100% pick out abnormal grains such as low brightness, high voltage, damage, short edge, blue mark, etc., which leads to the flow of abnormal grains to the client and customer complaints.

[0003] At the same time, when inspecting the appearance of grains through AOI, due to the limited technology of AOI machines, it is easy to miss inspections, and it is impossible to output the appearance data of all grains, let alone small-sized grains with abnormal appearance. It can only rely on post-process manual visual inspection to remove abnormal grains through visual inspection, which is time-consuming, labor-intensive, and inefficient. In addition, manual suction cannot be done thoroughly, which will cause secondary damage to the product.

[0004] Therefore, how to improve the sorting accuracy of Mini LED chips has become a problem that needs to be solved urgently. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a Mini LED chip sorting method and system, which can achieve effective card control of abnormal grains and improve the factory quality of products.

[0006] In order to solve the above technical problems, the present invention provides a Mini LED chip sorting method, comprising: sampling the grains in a wafer to generate test data; filling the blank data formed by the sampling according to the test data, so that all the grains in the wafer have corresponding test data; cutting the wafer to cut the grains in the wafer into chips; expanding the wafer to expand the intervals between the chips; performing AOI inspection on the wafer to generate inspection data; sorting the chips in the wafer according to the test data and the inspection data to classify the chips into corresponding grades; and visually inspecting the wafer to identify appearance defects of the sorted chips.

[0007] As an improvement of the above scheme, the step of sampling the grains in the wafer includes: constructing a rectangular plane coordinate system; assigning coordinates to the grains in the wafer; and sampling the grains at equal intervals along the horizontal and vertical coordinate directions for point measurement.

[0008] As an improvement of the above scheme, the step of filling the blank data formed by the sampling according to the test data includes: taking the grains that have been spot-measured as reference grains, and taking the grains that have not been spot-measured as unmeasured grains; calculating the test data of the unmeasured grains between the two adjacent reference grains based on the test data of the two adjacent reference grains; and updating the unmeasured grains with the calculated test data as reference grains.

[0009] As an improvement of the above-mentioned scheme, the step of calculating the test data of unmeasured grains between two adjacent reference grains based on the test data of two adjacent reference grains includes: extracting the test data of two adjacent reference grains; obtaining the number of unmeasured grains between two adjacent reference grains; calculating the test data of unmeasured grains between two adjacent reference grains according to the formula Qi=Qs+i[(Qe-Qs) / (N+1)]; wherein Qs is the test data of the starting reference grain between the two adjacent reference grains, Qe is the test data of the last reference grain between the two adjacent reference grains, N is the number of unmeasured grains between the two adjacent reference grains, and Qi is the test data of the i-th unmeasured grain between the two adjacent reference grains.

[0010] As an improvement of the above solution, the test data includes electrical data and optical data.

[0011] As an improvement of the above-mentioned solution, the Mini LED chip sorting method also includes: combining the test data and the detection data into combined file data.

[0012] Correspondingly, the present invention also provides a Mini LED chip sorting system, comprising: a spot testing machine, used to perform spot testing on the grains in the wafer to generate test data; a main control machine, used to fill the blank data formed by the spot testing according to the test data, so that all the grains in the wafer have corresponding test data; a cutting machine, used to cut the wafer to cut the grains in the wafer into chips; an expanding machine, used to expand the wafer to expand the interval between the chips; an AOI inspection machine, used to perform AOI inspection on the wafer to generate inspection data; a sorting machine, used to sort the chips in the wafer according to the test data and the inspection data, so as to classify the chips into corresponding grades; a visual inspection machine, used to perform visual inspection on the wafer to identify the appearance defects of the sorted chips.

[0013] As an improvement of the above scheme, the point measurement machine includes: a coordinate system construction module for constructing a rectangular plane coordinate system; a coordinate allocation module for allocating coordinates to the grains in the wafer; and a point measurement module for extracting grains equidistantly along the horizontal and vertical coordinate directions for point measurement.

[0014] As an improvement of the above scheme, the main control machine includes: a setting module, which is used to use the grains that have been spot-measured as reference grains, and the grains that have not been spot-measured as unmeasured grains; a calculation module, which is used to calculate the test data of the unmeasured grains between two adjacent reference grains based on the test data of the two adjacent reference grains; and an update module, which is used to update the unmeasured grains with the calculated test data to the reference grains.

[0015] As an improvement of the above scheme, the calculation module includes: an extraction unit, used to extract test data of two adjacent reference grains; an acquisition unit, used to acquire the number of unmeasured grains between two adjacent reference grains; a calculation unit, used to calculate the test data of unmeasured grains between two adjacent reference grains according to the formula Qi=Qs+i[(Qe-Qs) / (N+1)]; wherein Qs is the test data of the starting reference grain between the two adjacent reference grains, Qe is the test data of the last reference grain between the two adjacent reference grains, N is the number of unmeasured grains between the two adjacent reference grains, and Qi is the test data of the i-th unmeasured grain between the two adjacent reference grains.

[0016] The implementation of the present invention has the following beneficial effects:

[0017] The present invention obtains sampling data by sampling grains, and estimates full measurement data according to the characteristics of the sampling data and the distribution law of the sampled grains, so that in subsequent sorting, defective products can be sorted out through point-to-point data of coordinates; thereby, effective card control of abnormal grains can be achieved, abnormal grains can be stuck in the factory, the abnormal PPM value of grains can be reduced, customer complaints can be reduced, and product quality can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flow chart of an embodiment of a Mini LED chip sorting method of the present invention;

[0019] Figure 2 It is a schematic diagram of test data of a wafer sampled and tested in the Mini LED chip sorting method of the present invention;

[0020] Figure 3 It is a schematic diagram of test data of all wafers in the Mini LED chip sorting method of the present invention;

[0021] Figure 4 It is a schematic diagram of the detection data of all wafers in the Mini LED chip sorting method of the present invention;

[0022] Figure 5 It is a schematic diagram of combined data of all wafers in the Mini LED chip sorting method of the present invention;

[0023] Figure 6It is a schematic diagram of the structure of an embodiment of the Mini LED chip sorting system of the present invention. DETAILED DESCRIPTION

[0024] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings.

[0025] See also Figure 1 , Figure 1 The flowchart of an embodiment of the Mini LED chip sorting method of the present invention is shown, which includes:

[0026] S101, sampling and testing the dies in the wafer to generate test data;

[0027] like Figure 2 As shown, the steps of sampling the dies in the wafer include:

[0028] (1) Construct a rectangular plane coordinate system;

[0029] (2) assigning coordinates to the grains in the wafer;

[0030] Specifically, a polar coordinate system or equations may be used to generate coordinates of all points on the wafer.

[0031] (3) Grains are equidistantly selected along the horizontal and vertical axes for point measurement to generate test data.

[0032] It should be noted that the sampling interval can be freely set according to actual conditions.

[0033] For example, one grain out of every 10 grains along the horizontal axis direction may be selected for point measurement;

[0034] For another example, one grain out of every 10 grains along the ordinate direction may be selected for point measurement;

[0035] For another example, one grain out of every 10 grains along the horizontal axis direction may be selected for point measurement, and one grain out of every 10 grains along the vertical axis direction may be selected for point measurement.

[0036] In practical applications, the dies in the wafer can be sampled by a spot tester to obtain test data of some dies; wherein the test data includes electrical data and optical data.

[0037] S102, fill the blank data formed by sampling according to the test data, so that all the dies in the wafer correspond to the test data (see Figure 3 );

[0038] It should be noted that filling the gap data means that according to the test data of the grains that have been spot-tested, the test data is assigned to the nearby grains that have not been spot-tested, so that all the grains in the wafer have corresponding test data. Among them, the electrical data includes forward voltage (VF), reverse current (IR), etc., and the optical data includes brightness (LOP), dominant wavelength (WLD), etc., but it is not limited to this and can be adjusted according to actual conditions.

[0039] Accordingly, the step of filling the gap data formed by the random test according to the test data includes:

[0040] (1) The grains that have been spot-measured are regarded as reference grains, and the grains that have not been spot-measured are regarded as unmeasured grains;

[0041] (2) calculating the test data of the untested grains between the two adjacent reference grains based on the test data of the two adjacent reference grains;

[0042] Specifically, the step of calculating the test data of the untested die between the two adjacent reference die according to the test data of the two adjacent reference die comprises:

[0043] (2.1) Extracting test data of two adjacent reference grains;

[0044] (2.2) obtaining the number of unmeasured grains between two adjacent reference grains;

[0045] (2.3) According to the following formula, calculate the test data of the untested grains between two adjacent reference grains:

[0046] Qi=Qs+i[(Qe-Qs) / (N+1)]

[0047] in,

[0048] Qs is the test data of the starting reference grain among two adjacent reference grains;

[0049] Qe is the test data of the last reference grain among two adjacent reference grains;

[0050] N is the number of unmeasured grains between two adjacent reference grains;

[0051] Qi is the test data of the i-th untested die between two adjacent reference die.

[0052] For example, in the same row, there are unmeasured grains A2, A3 and A4 between the adjacent reference grains A1 and A5; wherein the dominant wavelength of the reference grain A1 is 655nm, and the dominant wavelength of the reference grain A5 is 660nm, then it can be calculated that the dominant wavelength of the unmeasured grain A2 is Q1=655+[(660-655) / (3+1)]=656.25nm, the dominant wavelength of the unmeasured grain A3 is Q2=655+2[(660-655) / (3+1)]=657.5nm, and the dominant wavelength of the unmeasured grain A4 is Q3=655+3[(660-655) / (3+1)]=658.75nm.

[0053] (3) The unmeasured grains of the calculated test data are updated as the reference grains.

[0054] After the unmeasured grains with calculated test data are updated as reference grains, the unmeasured grains can continue to be assigned values ​​by using the test data of the latest reference grain as a reference until all grains have completed the assignment of values.

[0055] S103, cutting the wafer to cut the grains in the wafer into chips;

[0056] The die pattern on the wafer is cut into individual chips using two-knife dicing technology.

[0057] S104, performing expansion processing on the wafer to enlarge the interval between the chips;

[0058] Expansion usually refers to the process of increasing the chip spacing on a wafer through specific technical means during the production process to facilitate subsequent process operations; this process is of great significance for improving production efficiency and reducing material waste.

[0059] After the wafer is cut, the chips are closely arranged and the spacing is very small (about 0.1mm), which is not conducive to the subsequent process operation. Therefore, the film of the wafer is expanded using a wafer expansion machine to stretch the chip spacing to about 0.6mm, which can facilitate the subsequent appearance inspection and grain sorting operations.

[0060] S105, performing AOI inspection on the wafer to generate inspection data;

[0061] AOI (Automatic Optical Inspection) uses automatic optical inspection technology to identify and classify defects on wafers.

[0062] Due to the small size of the product, the AOI machine has many missed scanning problems, resulting in some abnormal grains being unable to be identified. Figure 4 As shown, the dot-shaped areas in the wafer are the grains that were missed.

[0063] Furthermore, the present invention combines the test data and the detection data into a combined file data (see Figure 5 ) to ensure that each chip has corresponding archive data, providing data support for subsequent sorting processing.

[0064] S106, sorting the chips in the wafer according to the test data and the detection data to classify the chips into corresponding grades;

[0065] Chip sorting is a process of classifying crystal grains, which involves classifying the chips on the wafer. The present invention classifies the chips on the wafer into different grades according to the test data and the appearance inspection results (ie, the detection data).

[0066] For example, some chips may be classified as high-performance chips, while others may be classified as inferior due to certain defects.

[0067] Different from the prior art, each chip in the present invention has corresponding test data. When the detection data is missed, the chips can be forcibly sorted according to the test data, thereby ensuring the sorting effect.

[0068] S107, performing visual inspection on the wafer to identify appearance defects of the sorted chips.

[0069] Visual inspection is the use of manual inspection under a microscope to identify the appearance defects of the sorted chips to ensure that the quality and performance of the chips meet customer requirements. Only chips that pass the inspection can be put into storage.

[0070] Therefore, the present invention obtains sampling data by sampling grains, and estimates full measurement data according to the characteristics of the sampling data and the distribution law of the sampled grains, so that in subsequent sorting, defective products can be sorted out by point-to-point data of coordinates; thereby, effective control of abnormal grains can be achieved, abnormal grains can be stuck in the factory, the abnormal PPM value of grains can be reduced, customer complaints can be reduced, and product quality can be improved.

[0071] See also Figure 6 , Figure 6 The specific structure of the Mini LED chip sorting system 100 of the present invention is shown, which includes:

[0072] The spot tester 1 is used for sampling the dies in the wafer to generate test data.

[0073] The main control machine 2 is used to fill the blank data formed by the sampling test according to the test data, so that all the grains in the wafer have corresponding test data; wherein, filling the blank data means that according to the test data of the grains that have been spot-tested, the test data is assigned to the nearby grains that have not been spot-tested, so that all the grains in the wafer have corresponding test data.

[0074] The cutting machine 3 is used to cut the wafer to cut the grains in the wafer into chips; preferably, a two-knife cutting technology can be used to cut the grain pattern on the wafer into individual chips.

[0075] The wafer expansion machine 4 is used to expand the wafer to enlarge the interval between chips. By using the wafer expansion machine to expand the film bonding the wafer, the interval between chips can be stretched from 0.1mm to about 0.6mm, which can facilitate subsequent appearance inspection and grain sorting operations.

[0076] AOI inspection machine 5, used to perform AOI (Automatic Optical Inspection) inspection on the wafer to generate inspection data;

[0077] The sorting machine 6 is used to sort the chips in the wafer according to the test data and the detection data so as to classify the chips into corresponding grades. Chip sorting is a process of classifying grains, which involves classifying the chips on the wafer. The present invention classifies the chips on the wafer into different grades according to the test data and the appearance inspection results (i.e., the detection data).

[0078] The visual inspection machine 7 is used to perform visual inspection on the wafers to identify the appearance defects of the sorted chips. Visual inspection is to use manual inspection under a microscope to identify the appearance defects of the sorted chips to ensure that the quality and performance of the chips meet customer requirements. The chips can only be put into storage after passing the inspection.

[0079] It should be noted that due to the small size of the product, the AOI machine has many missed scan problems, resulting in the inability to identify some abnormal grains. The present invention combines the test data and the detection data into combined data to ensure that each chip has corresponding combined data. When the detection data is missed, the chip can be forced to be sorted according to the test data, providing data support for subsequent sorting processing.

[0080] Furthermore, the spot measuring machine 1 comprises:

[0081] Coordinate system building module, used to build rectangular plane coordinate system.

[0082] The coordinate allocation module is used to allocate coordinates to the dies in the wafer; specifically, a polar coordinate system or equation can be used to generate the coordinates of all points on the wafer.

[0083] The point measurement module is used to extract grains at equal distances along the horizontal and vertical axes for point measurement; the measurement interval can be freely set according to actual conditions.

[0084] In practical applications, the grains in the wafer can be sampled through a spot test machine to obtain test data of some grains; the test data includes electrical data and optical data. The electrical data includes forward voltage (VF), reverse current (IR), etc. The optical data includes brightness (LOP), dominant wavelength (WLD), etc., but it is not limited to this and can be adjusted according to actual conditions.

[0085] In addition, the main control machine 2 includes:

[0086] A setting module, used for taking the grains that have been spot-measured as reference grains, and taking the grains that have not been spot-measured as unmeasured grains;

[0087] A calculation module, used for calculating the test data of the untested grains between the two adjacent reference grains according to the test data of the two adjacent reference grains;

[0088] The updating module is used to update the untested grains of the calculated test data to the reference grains.

[0089] Therefore, after the untested grains with calculated test data are updated as reference grains, the untested grains can continue to be assigned values ​​by using the test data of the latest reference grain as a reference until all grains have completed the assignment of values.

[0090] More preferably, the computing module comprises:

[0091] An extraction unit, used for extracting test data of two adjacent reference grains;

[0092] An acquisition unit, used for acquiring the number of unmeasured grains between two adjacent reference grains;

[0093] The calculation unit is used to calculate the test data of the unmeasured grains between two adjacent reference grains according to the formula Qi=Qs+i[(Qe-Qs) / (N+1)]; wherein Qs is the test data of the starting reference grain between the two adjacent reference grains, Qe is the test data of the last reference grain between the two adjacent reference grains, N is the number of unmeasured grains between the two adjacent reference grains, and Qi is the test data of the i-th unmeasured grain between the two adjacent reference grains.

[0094] As can be seen from the above, the present invention has the following beneficial effects:

[0095] 1. Improve product quality, ensure that all chips shipped out of the factory undergo strict quality inspection, reduce or eliminate the outflow of defective products, and thus improve overall product quality;

[0096] 2. Enhance customer trust by providing defect-free, high-quality products to increase customer trust and satisfaction with the brand, thereby improving the corporate image;

[0097] 3. Reduce after-sales costs, reduce after-sales service needs caused by quality problems, and reduce the cost of after-sales repairs and replacements.

[0098] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A Mini LED chip sorting method, characterized in that: include: Perform sampling test on the dies in the wafer to generate test data; Filling the blank data formed by sampling according to the test data, so that all the dies in the wafer correspond to the test data; Performing a dicing process on the wafer to cut the dies in the wafer into chips; Performing an expansion process on the wafer to enlarge the interval between the chips; Performing AOI inspection on the wafer to generate inspection data; Sorting the chips in the wafer according to the test data and the detection data to classify the chips into corresponding grades; The wafer is visually inspected to identify appearance defects of the sorted chips.

2. The Mini LED chip sorting method according to claim 1, characterized in that: The step of sampling and testing the dies in the wafer comprises: Construct a rectangular plane coordinate system; assigning coordinates to dies in the wafer; Grains are equally spaced along the horizontal and vertical directions for point measurement.

3. The Mini LED chip sorting method according to claim 2, characterized in that: The step of filling the gap data formed by the random test according to the test data comprises: The grains that have been spot-measured are taken as reference grains, and the grains that have not been spot-measured are taken as unmeasured grains; Calculating the test data of the untested grains between the two adjacent reference grains according to the test data of the two adjacent reference grains; Update the unmeasured die of the calculated test data to the reference die.

4. The Mini LED chip sorting method according to claim 3, characterized in that: The step of calculating the test data of the untested grains between the two adjacent reference grains according to the test data of the two adjacent reference grains comprises: Extract test data of two adjacent reference grains; Obtaining the number of unmeasured grains between two adjacent reference grains; According to the formula Qi=Qs+i[(Qe-Qs) / (N+1)], the test data of the untested grains between two adjacent reference grains are calculated; Among them, Qs is the test data of the starting reference grain between two adjacent reference grains, Qe is the test data of the last reference grain between two adjacent reference grains, N is the number of unmeasured grains between two adjacent reference grains, and Qi is the test data of the i-th unmeasured grain between two adjacent reference grains.

5. The Mini LED chip sorting method according to claim 1, wherein: The test data includes electrical data and optical data.

6. The Mini LED chip sorting method according to claim 1, wherein: Also includes: The test data and the detection data are combined into archived data.

7. A Mini LED chip sorting system, characterized in that: include: Spot test machine, used to sample the dies in the wafer to generate test data; A main control machine, used for filling the blank data formed by sampling according to the test data, so that all the dies in the wafer correspond to the test data; A cutting machine, used for cutting the wafer to cut the grains in the wafer into chips; A wafer expansion machine, used for expanding the wafer to enlarge the interval between the chips; An AO I inspection machine, used for performing AO I inspection on the wafer to generate inspection data; A sorting machine, used for sorting the chips in the wafer according to the test data and the detection data, so as to classify the chips into corresponding grades; The visual inspection machine is used to perform visual inspection on the wafer to identify appearance defects of the chips after sorting.

8. The Mini LED chip sorting system according to claim 7, characterized in that: The point measuring machine comprises: A coordinate system building module, used to build a rectangular plane coordinate system; A coordinate allocation module, used for allocating coordinates to the dies in the wafer; The point measurement module is used to extract grains at equal distances along the horizontal and vertical axes for point measurement.

9. The Mini LED chip sorting system according to claim 7, characterized in that: The main control machine includes: A setting module, used for taking the grains that have been spot-measured as reference grains, and taking the grains that have not been spot-measured as unmeasured grains; A calculation module, used for calculating the test data of the untested grains between the two adjacent reference grains according to the test data of the two adjacent reference grains; The updating module is used to update the untested grains of the calculated test data to the reference grains.

10. The Mini LED chip sorting system according to claim 9, wherein: The calculation module comprises: An extraction unit, used for extracting test data of two adjacent reference grains; An acquisition unit, used for acquiring the number of unmeasured grains between two adjacent reference grains; A calculation unit, used to calculate the test data of the untested grains between two adjacent reference grains according to the formula Qi=Qs+i[(Qe-Qs) / (N+1)]; Among them, Qs is the test data of the starting reference grain between two adjacent reference grains, Qe is the test data of the last reference grain between two adjacent reference grains, N is the number of unmeasured grains between two adjacent reference grains, and Qi is the test data of the i-th unmeasured grain between two adjacent reference grains.

Images